January 15th, 2018 January 17, 2018

0 downloads 0 Views 5MB Size Report
Jan 15, 2018 - has become particularly significant over the past 10 years with the development ..... the organization, the learning construct must be group and not just ...... These skills read more like vague clusters of abilities or even personality traits ...... Retrieved from https://hbr.org/2008/07/the-competitive-imperative-of-.
Stanley J. Smits, Department of Managerial Sciences, Robinson College of Business, Georgia State University, Atlanta, GA,USA

January 15th, 2018

JC Spender, Visiting Scholar at Rutgers Business School,Newark, New Jersey, USA

January 17, 2018

Strategic Value of Tacit Knowledge and Learning Processes in a Young E-commerce Company

Dissertation Submitted to the International School of Management

in Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY International Business Management

by MATTHEW A. ANDREWS Paris, France December 2017

i

Abstract This study explores the links among tacit knowledge, organizational learning, and competitive advantage in a young e-commerce company using a mixed methods case study approach. Technology companies tend to focus on technical skills and abilities rather than on processes for learning, and sharing knowledge whereas a company’s tacit knowledge, embedded in its routines and practices represent firm-specific assets which are potential competitive advantages for the company’s long-term growth. Moreover, there are few studies which address these issues in young technology companies. The purpose of this research is to assess the tacit knowledge possessed by employees of a young e-commerce firm to help identify critical organizational skills; to explore ways of generating new explicit knowledge from this tacit knowledge through practice so that the company may leverage its organizational skills and knowledge as sustainable competitive advantages; and to assess the depth of learning in the company with a view to identifying areas of strength and weakness. The theoretical framework for this study combines elements from three fields in management and social sciences literature which contribute to our understanding of organizational knowledge and knowledge creation: organizational learning, learning organization, and knowledge management. The empirical research carried out for this study solicited the participation of 42 company representatives (36% of the total population) and made use of the Learning Organization Survey, a quantitative multiple choice survey instrument developed by Garvin, Edmondson, and Gino (2008), as well as a semi-structured interview method based on the causal mapping approach proposed by Ambrosini and Bowman (2001), and based on the knowledge management framework developed by McIver, Lengnick-Hall, Lengnick-Hall, and Ramachandran (2013).

ii

Findings from the survey suggest the company has strengths when it comes to a supportive learning environment but weaknesses when it comes to concrete learning processes and practices. Results from the semi-structured interviews largely confirm the survey results, and they provide context for a richer interpretation of those results. The interviews also revealed a number of tacit skills embedded in company practices and routines which are related to key organizational constructs such as innovation, commercial success, and employee/client autonomy. An assessment of several key company positions using the knowledge management framework revealed more company tacit skills embedded in activities which straddle several positions, and embedded in how employees interact, rather than inherent to the individual positions themselves. Recommendations to the company include implementing tools to encourage socialization and knowledge sharing, as well as implementing more effective overall knowledge management practices and policies. Recommendations to help e-commerce companies, in general, develop long-term sustainable competitive advantage include: investing in learning processes and knowledge management tools; embedding learning in the workplace; and developing moderately complex strategies whereby technology is intertwined with practice, tacit skills and knowledge. Some promising areas for future research include exploring links between organizational life cycle stages of development and organizational learning; assessing the role of communities of practice in e-commerce companies with respect to learning; evaluating the transferability of knowledge in e-commerce companies and the related issues of complexity and knowledge leaks.

iii

Acknowledgements I would like to express my gratitude to my dissertation supervisor, Dr. Stan Smits, who provided effective guidance throughout the process of writing this dissertation. The breadth of his knowledge concerning strategy, change management, and organizational life cycles contributed enormously to my wider understanding of the importance of tacit knowledge. Dr. Smits has a deep understanding of the value of research involving close work with practitioners and which requires spending time in organizations to gather data. On that note, I would also like to express my gratitude to the study company, particularly the e-commerce manager, and the human resources manager, without whose support and trust this dissertation would not have been possible. As many researchers know, getting companies to cooperate with research is difficult especially when significant numbers of employees are involved. There are numerous potential obstacles which range from internal resistance to the way the project may interfere with production schedules, or simply the extra work and time involved in scheduling and participating in the interviews. In company research requires champions within the organization who are committed to the project in spite of these obstacles.

iv

Table of Contents Chapter 1: Introduction ................................................................................................................... 1 Statement of the Problem .................................................................................................... 1 Purpose of the Study ........................................................................................................... 2 Significance of the Study .................................................................................................... 2 Theoretical Framework ....................................................................................................... 3 The Present Study ............................................................................................................... 4 Assumptions........................................................................................................................ 6 Limitations .......................................................................................................................... 6 Research Questions ............................................................................................................. 7 Chapter 2: Literature Review .......................................................................................................... 9 Introduction to the E-commerce Industry ........................................................................... 9 Organizational Learning, Learning Organization, and Knowledge Management ............ 28 Theoretical Framework for the Study ............................................................................... 40 Chapter 3: Research Design and Methodology ............................................................................ 43 Purpose of the Study ......................................................................................................... 43 Description of the Company and Population Being Studied ............................................ 44 Design Overview .............................................................................................................. 50 Specific Design Elements ................................................................................................. 62 Data Gathering .................................................................................................................. 79 Data Analysis .................................................................................................................... 84 Chapter 4: Results ......................................................................................................................... 90 Introduction ....................................................................................................................... 90 Survey Results .................................................................................................................. 90 Results from Semi-structured Interviews........................................................................ 102 Answers to Research Questions ...................................................................................... 133 Summary of Major Findings ........................................................................................... 143 Chapter 5: Discussion, Recommendations, and Conclusions ..................................................... 148 Synopsis .......................................................................................................................... 148 Discussion ....................................................................................................................... 159 Concluding Remarks ....................................................................................................... 211 References ................................................................................................................................... 213 Appendices .................................................................................................................................. 213 Appendix A1: The Learning Organization Survey (English version) ............................ 226 Appendix A2: The Learning Organization Survey (French version) ............................. 229 Appendix B: KIP, Performance, KM Strategies, and Types of Work ............................ 232 Appendix C1: Individual Interviews ............................................................................... 233 Appendix C2: Group Interviews ..................................................................................... 246 Appendix D: Complete Scoring Data Sheet for Learning Organization Survey ............ 263

v

Appendix E: Complete Breakdown of Benchmark Scores Compiled by the Survey Authors ..................................................................................................................... 265 Appendix F: Table of Combined Survey Results ........................................................... 266 Appendix G: Tables of Survey Results by Department and by Level of Hierarchy....... 268 Appendix H: Five Number Summaries, Box Plots, and Histograms of Survey Results 279 Appendix I: Causal Mapping of Individual Positions..................................................... 291 Appendix J: KIP Framework Applied to Specific Company Positions .......................... 300

vi

List of Tables Table 1. Sequence of initial interviews and formal survey/interview sessions ............................ 83 Table 2. Comparison of medians provided by survey authors to company medians ................... 92 Table 3. Comparison of medians provided by survey authors to company means....................... 92 Table 4. Differences in median-to-median comparison, and median-to-mean comparison ......... 93 Table 5. Benchmark scores for the Learning Organization Survey .............................................. 94 Table 6. Benchmark comparison for the combined company results for the Supportive Learning Environment building block ............................................................................ 95 Table 7. Benchmark comparison for the combined company results for the Concrete Learning Processes building block................................................................................................. 96 Table 8. Benchmark comparison for the combined company results for the Leadership that Reinforces Learning building block ............................................................................... 97 Table 9. Standard deviation for the combined survey items ......................................................... 98 Table 10. Survey results across departments ................................................................................ 99 Table 11. Survey results across levels of hierarchy .................................................................... 100 Table 12. Tacit skills revealed during semi-structured interviews analyzed according to Crossan et al.’s (1999) learning processes ................................................................. 171

vii

List of Figures Figure 1. Degrees of tacitness ....................................................................................................... 70 Figure 2. Researching tacit skills: a summary of the proposed method ....................................... 73 Figure 3. Knowledge-in-practice for organizational work ........................................................... 75 Figure 4. Knowledge-in-practice framework questions................................................................ 77 Figure 5. Integrating the knowledge-in-practice framework into the causal mapping methodology ................................................................................................................. 78 Figure 6. Causal mapping of key organizational construct, innovation ..................................... 103 Figure 7. Causal mapping of key organizational construct: commercial success....................... 108 Figure 8. Causal mapping of key organizational success factor: good relations with the clients.......................................................................................................................... 111 Figure 9. Causal mapping or organizational success factor: client and employee development ............................................................................................................... 115 Figure 10. Knowledge-in-practice framework ............................................................................ 121 Figure 11. Knowledge-in-practice analyses of after sales service activity ................................. 122 Figure 12. Knowledge-in-practice analyses for senior e-store managers ................................... 124 Figure 13. Knowledge-in-practice analyses for lead IT architect, and the supply chain logistics manager ........................................................................................................ 125 Figure 14. Knowledge-in-practice analyses for an embedded e-commerce position ................. 126 Figure 15. Knowledge-in-practice analysis for e-commerce activity: head of traffic and CRM, and brand manager positions ........................................................................... 128 Figure 16. Knowledge-in-practice analysis for co-founder/head of photography ...................... 129 Figure 17. Knowledge-in-practice diagram with key job positions ............................................ 130 Figure 18. Knowledge-in-practice diagram with key activities .................................................. 131 Figure 19. Organizational stages of development....................................................................... 162 Figure 20. Integrating political strategies into the 4I learning processes framework ................. 175 Figure 21. Knowledge generation, and learning processes......................................................... 176 Figure 22. Learning processes and degrees of tacitness ............................................................. 177

viii

Chapter 1: Introduction Statement of the Problem E-commerce is a sector that is experiencing rapid growth and is having a major impact on how consumers shop, how companies manage their branding and communications, and how transactions and deliveries are made. Young, successful B2B companies have been able to take advantage of the opportunities stemming from this transformation by offering technical knowhow and services which other companies need to stay competitive. Many of the employees of these companies are young, have grown up with the Internet, are technically trained, but relatively inexperienced. For young companies in this sector experiencing fast growth there are inherent management challenges. Such companies, however, tend to focus on technical skills and abilities rather than on processes for learning and sharing knowledge. Meanwhile a company’s tacit knowledge, embedded in its routines and practices represent firm-specific assets which are potential competitive advantages for the company’s long-term growth. Results of this study may be used by managers and scholars to better understand how the unique capabilities of an enterprise to transform inputs into outputs with greater value are embedded in its processes and values. Furthermore, the results will provide insights about the relationships which exist among a firm’s tacit knowledge, its explicit knowledge, its ability to create new knowledge and the organization’s culture of learning and knowledge sharing. This will have implications for management practices in the e-commerce sector because processes, learning, knowledge, and knowledge sharing will be examined for their strategic importance rather than technology and technical skills.

1

Purpose of the Study The purpose of this study is to assess the tacit knowledge possessed by employees of a young e-commerce firm to help identify critical organizational skills; to explore ways of generating new explicit knowledge from this tacit knowledge through practice so that the company may leverage its organizational skills and knowledge as sustainable competitive advantages; and to assess the depth of learning in the company with a view to identifying areas of strength and weakness. As identified in the Statement of the Problem, companies in this sector do not give adequate attention to such questions. Moreover, there are few studies involving young technology companies which focus on tacit knowledge, knowledge creation, or on organizational learning. Managers should be more aware of the strategic importance of such areas when they attempt to assess the unique advantages of their companies, and when conceiving of and implementing long-term strategic plans. Christensen (1997/2011) argues that an organization’s processes and values are more important than its resources especially concerning innovations: Indeed, we could deal identical sets of resources to two different organizations, and what they created from those resources would likely be very different—because the capabilities to transform inputs into goods and services of greater value reside in the organization’s processes and values” (1997/2011, p. 187). This study will provide an example of how to assess an e-commerce company’s processes and values and will have implications for how to leverage the insights gained for strategic value. Significance of the Study Despite much interest in the topic of tacit knowledge and its strategic importance to the firm (Grant, 2013; Lazar & Robu, 2015; Nonaka, 1994; Spender 1993, 2014), there is a need to know more empirically about the construct for it to become coherent and convincing (Ambrosini

2

& Bowman, 2001). Argote (2005) proposes that a better understanding of processes of organizational learning will help researchers and practitioners design more effective knowledge management systems to “capture and transfer knowledge acquired through that learning” (p. 46). Furthermore, even though it is widely accepted that an organization’s ability to create and manage knowledge is critical for its sustainability (Bettis & Hitt, 1995), McIver, Lengnick-Hall, Lengnick-Hall, and Ramachandran (2013) state that “our understanding of how work has changed and how to organize, manage, and integrate activities for managing knowledge that are essential to work activities is limited” (p. 597). These authors further note a dearth of theories to explain how knowledge management activities lead to desired outcomes, and a lack of discussions which attempt to apply knowledge management activities in today’s transformed work environment (McIver et al., 2013). Also, as noted in the Purpose of the Study, there are few published studies which focus on tacit knowledge, knowledge management or organizational learning in young technology companies. This study will attempt to fill those gaps in the literature; as a scientific contribution this study will produce a model for assessing tacit knowledge and the effectiveness of knowledge management practices in a young technology company. Furthermore, it will assess the extent to which employees in the company perceive it to be a learning organization; the results of this assessment will be compared across company departments as well as compared to other companies. Theoretical Framework The theoretical framework for this study combines elements from three fields in management and social sciences literature which contribute to our understanding of organizational knowledge and knowledge creation: organizational learning, learning organization, and knowledge management. While the field of organizational learning is

3

considered more academically robust compared to the field of the learning organization, which has been criticized for being fragmented, theoretically flawed, and utopian, this study combines elements of both fields and recognizes that they are closely related. For example, “double-loop learning” (Argyris, 2002), developed in the field of organizational learning, refers to transformative learning in which underlying values are questioned and/or changed. Senge (1990/2006) builds on this notion in his concept of the learning organization, and Garvin (1993/2000) proposes a framework for making the learning organization actionable. This actionable framework forms the basis for the Learning Organization Survey developed by Garvin, Edmondson, and Gino (2008). This survey was administered to the employees of the ecommerce company and the results can be compared across departments within the company and also compared to the benchmark scores provided by the authors of the survey. The tool is designed to assess five main areas of organizational learning and can therefore help the enterprise identify areas which need to be strengthened—this in turn can lead to concrete recommendations. . The Present Study As previously stated, the purpose of this study is threefold. Firstly, it is to assess the depth of learning in a young French e-commerce company with a view to identifying areas of strength and weakness. Secondly, it is to assess the tacit knowledge possessed by this firm in order to help identify critical organizational skills. Finally, it is to explore ways of generating new explicit knowledge from this tacit knowledge through practice so that the company may leverage its organizational skills and knowledge as sustainable competitive advantages. The empirical research carried out for this study solicited the participation of 42 company representatives (slightly less than half of the total personnel) who represented all levels of the

4

hierarchy—the directors, the middle managers and the technical employees. All participants completed the Learning Organization Survey (Garvin et al., 2008), a multiple choice survey instrument described in the Theoretical Framework which uses a seven- or five- point Likert scale; each survey was completed on paper immediately preceding each interview. Five company directors were interviewed individually—these interviews lasted approximately 45 minutes. The remaining participants (middle managers and employees), were interviewed in five separate groups each consisting of from five to nine participants. Each group interview lasted approximately one hour. The method for carrying out these individual and group semi-structured interviews was developed from the work of Ambrosini and Bowman (2001) on causal mapping; the objective of these group interviews was to assess the tacit knowledge and to identify critical organizational skills. All the interviews took place over a 15-day period in the company and were conducted in French by the researcher who is fluent in French and English; the raw notes taken by hand by the researcher were then converted into interview transcripts in English. Finally, the results of the Learning Organization Survey (Garvin et al., 2008) were then assessed quantitatively and compared across departments as well as compared to the benchmark scores proposed by Garvin et al. (2008). This allowed the researcher to identify potential areas of strength and weakness according to the theory of the Learning Organization. The transcripts from the semi-structured interviews were analyzed qualitatively based on the method of causal mapping (Ambrosini & Bowman, 2001), and according to the knowledge management framework developed by McIver et al. (2013). Finally, an assessment of the survey results, and of the results of the semi-structured interviews were used as a basis from which to develop recommendations for how the company might generate new explicit knowledge, and how it might leverage its knowledge and skills as sustainable competitive advantages.

5

Assumptions 

The Learning Organization Survey (Garvin et al., 2008) is a valid survey instrument for the company studied notwithstanding the company’s specific cultural, demographic, technological, and business context.



The learning organization concept is a sufficiently coherent construct to provide a meaningful framework for assessing the learning processes of the company studied.



Data collected using the Learning Organization Survey can be aggregated on the department and company levels, compared across company departments and compared to the benchmark scores provided by Garvin et al. (2008) in a meaningful and valid manner to identify potential strengths and weaknesses in the company’s organizational learning processes and strategies.



Company tacit knowledge and learning can be effectively assessed through semistructured qualitative interviews.



The tacit knowledge and learning assessed through the interviews will provide insights which pertain to the company’s overall learning processes.



Tacit knowledge, and organizational learning, when properly understood and effectively managed are strategic assets to the company.

Limitations There are a number of limits to this study. Firstly, while this is a theory-driven study which will allow the findings to apply to some degree wherever the theory is applicable, the study focuses on one company. The applicability of the findings will therefore be limited especially where the industry, employees, context, and culture may be different. Secondly, the study made use of a nonrandom sampling technique. The management of the company solicited

6

the participation of all the employees, and scheduled the individual and group interviews. Participation, however, was voluntary—the final composition of the groups were influenced by availability and individual workload. While some may argue this lack of random, systematic sampling hinders the generality of the findings (Pinsonneault & Kramer, 1993), this is partly compensated by the high percentage of the population represented (approximately 38%), and by the diversity of hierarchical levels, and types of jobs held by the participants. Thirdly, as discussed more in Chapter 2: Literature Review, there are inherent challenges to conducting empirical research on tacit knowledge, knowledge creation, and organizational learning; and, there are theoretical weaknesses in the concept of the learning organization. There are also differing assumptions and hence lack of broad agreement about the nature of knowledge and how to study it; hence the methodology and findings of this study may be acceptable to some but not to all. Finally, some of the constructs used in this study are difficult to operationalize and to measure in a reliable and repeatable manner. Research Questions The research problems in this study are: a)

How are the concepts of organizational learning, knowledge management, and the learning organization related?

b)

What are different types of organizational knowledge and how are they related?

c)

How can firm-specific tacit knowledge of employees in an organization be mapped to help identify crucial organizational skills?

d)

What types of practices in the company contribute to the creation of new knowledge, and how might the company leverage those practices to develop new explicit

7

knowledge which can potentially be leveraged as a sustainable competitive advantage? e)

How can learning be assessed in an organization with a view to identifying strengths and weaknesses?

8

Chapter 2: Literature Review Introduction to the E-commerce Industry Definition of E-commerce. E-commerce is defined as “digitally enabled commercial transactions between and among organizations and individuals” (Laudon & Traver, 2017, p. 51). Transactions take place using the Internet, the Web, browsers, and mobile applications running on mobile devices. If such transactions and processes take place within a firm, however, they are not considered e-commerce but rather as e-business (Laudon & Traver, 2017). The two main types of e-commerce are business-to-consumer, and business-to-business. Business-to-consumer e-commerce, when companies try to reach individual consumers via the Internet, is primarily a retail model; whereas, business-to-business e-commerce, when companies engage with other companies via the Internet, is primarily a supply chain model concerned with logistics and procurement (Gereffi, 2001). Business-to-business e-commerce is far larger in value and volume than business-to-consumer e-commerce even though the latter is the most commonly discussed type. For example, as measured by the value of transactions in the United States, business-tobusiness e-commerce is estimated to be worth $6.3 trillion compared to $563 million for business-to-consumer e-commerce (Laudon & Traver, 2017, pgs. 58, 61); in the US and worldwide, it is estimated to be twice the size of the business-to-consumer market (Brown, 2015). However, ways of conducting e-commerce are far more varied than the above two categories would imply. Recent technological developments on the Internet and mobile devices have led to new ways to buy and sell online. For example, thanks to platforms such as eBay, consumers sell directly to other consumers via online auctions. Similarly, services such as easyCar Club or Drivy allow private, individual car owners to rent their vehicles to nearby

9

consumers. Moreover, Uber has exploited a mobile platform to link customers according to where they are located to drivers who use their own personal cars to offer rides similar to a taxi service (Petropolous, 2016). Platforms like eBay, Drivy, Airbnb, and Uber are sometimes referred to as online marketplaces because they process transactions for products and services provided by multiple third parties (Laumeister, 2014; Tozzi, 2008). In addition to selling directly to consumers, Amazon also runs its own online marketplace where a wide range of vendors sell products and services accessible through the Amazon website – these third party vendors are responsible for their own pricing and delivery. The advantage of online marketplaces is the amount of potential traffic attracted to the platform – often pricing is very competitive (Laumeister, 2014). The Web 2.0, which has facilitated user-generated content, has led to the rise of social media such as Facebook and Twitter. This development has also significantly impacted ecommerce because online merchants are placing more and more emphasis on reaching consumers, for example, by paying to have their ads shown to specific demographic groups (Merrill, 2016; Zaroban, 2015). But probably the biggest trend impacting e-commerce is the ever growing use of mobile devices such as smart phones, tablets and iPads to conduct transactions. Estimates put the number of smart phone users worldwide in 2016 at close to 2 billion, which represents a 12.6% increase over the prior year. According to eMarketer (“2 Billion Consumers Worldwide”, 2016), more than a quarter of the world’s population used smartphones by 2015, and more than a third of the world’s population (2.56 billion people) is expected to use one by 2018. Laudon and Traver (2016), consider the above mentioned developments and trends worthy of distinct categories of e-commerce which are: consumer-toconsumer e-commerce (like eBay), local e-commerce (like Uber), social e-commerce (facilitated

10

by social media like Facebook), and mobile e-commerce (conducted using mobile devices). However, I take the point of view that that these trends primarily represent variations of business-to-consumer e-commerce. E-commerce differentiated. E-commerce and the Internet, in general, are often associated with terms such as “revolutionary”, “disruptive” or “unprecedented” (Hoffman & Novak, 2000a, 2000b, Orlikowski & Iacono, 2000). Kelly (2005) argues, for example, that the Internet is akin to a giant planetary brain and suggests its development marks one of the most significant events in the history of humanity. Laudon and Traver (2016), note that the first few years of the Internet and e-commerce were characterized by utopian predictions of “friction free” commerce, “disintermediation”, and near total price transparency (p. 67). Shapiro and Varian (1999), meanwhile, suggest that the development of the Internet and e-commerce have much in common with the development of transportation infrastructure, and of earlier revolutionary technologies; they furthermore convincingly apply laws of classical economics to explain some of the most striking characteristics of the Internet and e-commerce. Similarly, Perez’ theory of Great Surges of Development (Hegedus, 2009; Perez, 2002, 2009a, 2009b) suggests that ecommerce is part of the fifth great surge in modern societies. Such a surge is defined as “the process of propagation of a technological revolution across the economy and society” (2009b, p. 780 footnote 2) and the fifth surge is referred to as “the age of information and digital communications” starting in the 1970’s (2009b, p. 782). The following paragraphs will attempt to describe some of the aspects of e-commerce that distinguish it from other forms of commerce and will examine the extent to which e-commerce defies the rules of traditional business and economics.

11

Despite their stance that the development of the Internet and e-commerce have parallels in history and are subject to the laws of economics, Shapiro and Varian (1999) articulate much of what is unique about them stating that, “the Internet, a hybrid between a broadcast medium and a point-to-point medium, offers exciting new potential for matching up customers and suppliers…web servers can observe the behavior of millions of customers and immediately produce customized content, bundled with customized ads” (p. 7). While the Internet and ecommerce have grown enormously in scope, volume, and sophistication since the publication of their book Information Rules (Shapiro & Varian, 1999), this description remains not only accurate and relevant today but is also one of the most concise summaries of the salient and distinctive features of the Internet and e-commerce. While many other authors and scholars have examined this question, few have so effectively and concisely captured the implications of ecommerce. Another significant contribution to this discussion is Evans and Wurster (1999). Defining the term reach as “the number of people, at home or at work exchanging information” (para. 7), they insist on the unprecedented potential reach of the Internet: “the rapid emergence of universal technical standards for communication, allowing everybody to communicate with everybody else at essentially zero cost, is a sea change” (para. 7). The authors furthermore insist on the potential for richness of the message in Internet communications; they define “richness” according to three aspects of the information being transmitted: bandwidth, how much the information can be customized, and the potential for interactivity. Most importantly, Evans and Wurster (1999) maintain that what had been accepted as a fundamental trade-off between richness and reach in the economics of information no longer applies in the era of e-commerce. Companies using the Internet today have the potential for nearly infinite reach, and a vast

12

richness of information. Underscoring their point, Laudon and Traver (2016) assert that the potential global reach of an e-commerce merchant is the entire online population estimated at 3.1 billion people in 2015 (p. 54). Levin (2011) attempts to address the question of what makes e-commerce distinctive by proposing three important unique aspects of e-commerce: scalability, customization, and innovation. While the customization aspect reaffirms one of the core themes already discussed above, his emphasis on scalability and innovation is noteworthy. To illustrate scalability he points out how many users platforms such as Facebook and Twitter are capable of handling compared to the low number of employees (particularly engineers) required to run the systems. He states that Facebook “grew to over 500 million users with less than 500 engineers, or one engineer for every million users” (p. 1). Levin furthermore insists on the link between customization and innovation; he suggests that given the relatively low cost of making changes to existing Internet platforms, and due to the emphasis on customizing the offer to the individual consumer, there are vast opportunities for continuous improvement in e-commerce, even if these opportunities are not always exploited. Bringing together many of the sources and ideas mentioned above, Laudon and Traver (2016) identify eight unique features of e-commerce: ubiquity, global reach, universal standards, richness, interactivity, information density, personalization/customization, and social technology. Ubiquity and global reach overlap because they refer to the continuous availability of the Internet to consumers in virtually any location where there is computer or mobile access, and they refer to the incredible potential reach this provides to e-merchants as already noted. Having such a vast amount of users able to communicate with each other is only possible thanks to universal standards which Internet systems and infrastructure have made possible. The fact that the

13

Internet is a mix between a broadcast and a point-to-point medium allows for the personalization/customization of the message and for high degrees of interactivity; furthermore the bandwidth allows for an unparalleled richness of information. The main contribution of Laudon and Traver to this discussion, other than synthesizing ideas and analyses of several different scholars and thinkers, is the addition of social technology. This feature of e-commerce has become particularly significant over the past 10 years with the development of the Web 2.0. Laudon and Traver state that “in a way quite different from all previous technologies, ecommerce technologies have evolved to be much more social by allowing users to create and share content with a worldwide community” (2016, p. 56). Clearly e-commerce is not just a new way of shopping or an incremental improvement in business efficiency thanks to digital technology. The economics of information has supplanted to a significant extent the dominance of the economics of manufacturing and inventory (Gereffi, 2001; Shaprio & Varian, 1999). This has involved changes in business organization, market structure and government regulations which are profound and lasting—every part of the global value chain is impacted, hence every activity necessary for the delivery of the final product. Moreover, new types of entities such as Google have risen to prominence—referred to as “infomediaries”—these entities control key entry points to the Internet and transform detailed information about consumer behavior into a highly valued asset (Gereffi, 2001; Shapiro & Varian, 1999). It is not surprising then that over the past 20 years, numerous technologies and business models have had to contend with new “disruptive technologies” (Christensen, 1997/2011). Fifteen out of 24 technologies listed in the introduction of The Innovator’s Dilemma (Christensen, 1997/2011) as having “disrupted” established technologies are Internet-based, are

14

related to e-commerce and/or can be considered forms of e-commerce. Some of these technologies include digital photography, mobile telephony, packet-switched communications networks, hand-held digital appliances, online stock brokering, online retailing, distance education, digital printing, and Internet protocols. In some cases, the new technology has almost entirely supplanted the older technology (as is the case with digital photography); in other cases, the new model coexists with the older model simultaneously competing with it, complementing it, and enriching it (as is the case with online retail versus brick-and-mortar retail). Past, present and future of e-commerce. The history of e-commerce involves multiple histories—the history of technological breakthroughs, of infrastructure development, of standards and regulations, of entrepreneurs and venture capitalists, and finally of evolving consumer behavior and consumer expectations. These multiple histories intertwine to form a sprawling, messy narrative, an in-depth discussion of which is beyond the scope of this work. This subsection attempts merely to outline a few of the most significant developments; it also examines business cycles, and explores some parallels with the histories of other technological innovations. The exact beginning of e-commerce is difficult to establish, and points of view differ on the topic. One of the earliest examples of someone shopping from home using electronic technology occurred in the UK in 1984, when a pensioner was able to order groceries from a local store using a television remote control thanks to a council initiative to help the elderly (Winterman & Kelly, 2013). Furthermore, in the same year, CompuServe opened the Electronic Mall which allowed customers using personal computer terminals to order general merchandise from 80 different participating firms (“CompuServe Unveils the ‘Electronic Mall”, 1984). A few years later, in 1989, Tim Berners-Lee of CERN first proposed the World Wide Web as a

15

“networked system for sharing information within the high-energy physics research community” (Orlikowski & Iacono, 2000, p. 359). One very early example of e-commerce using the World Wide Web is Book Stacks Unlimited, created in 1992 by Charles Stack—possibly the first online book seller (Rayport & Jaworski, 2004, p. 27). However, it was not until the mid-1990s that modern e-commerce really got its start with the launching of the Netscape Navigator Browser in 1994, and the lifting by the US National Sciences Foundation of the prohibition against commercial enterprises on the Internet the following year (Kelly, 2001; Orlikowski & Iacono, 2000). From this point on, the Internet became widely accessible, easy to use, and open for business. As previously mentioned, several scholars suggest that technological revolutions follow patterns and similar trajectories (Perez, 2009; Toffler, 1980; Varian, 2001). In 2001, Varian predicted that the Internet economy would pass through five stages similar to earlier technological innovations such as telegraphy, automobiles, and radio. The five stages he described include: experimentation, capitalization, management, hyper-competition, and consolidation. What follows is a brief summary of each stage. During the experimentation stage, entrepreneurs innovate with the evolving technology trying to find ways to create value and to discover new business models; then financiers and venture capitalists enter the picture in the capitalization stage because technological innovations generally require substantial up-front capital before they can generate revenue. Following this is the management stage when the focus is on building efficient operations, establishing brands and developing customer bases. In the next stage, hyper-competition, economies of scale dominate, and many firms disappear either because they lose financing or because they get acquired by larger firms. Finally, during the consolidation stage, a few big winners remaining from the hyper-competition stage concentrate

16

on standardizing their products, on fine-tuning their marketing, and on reducing their costs (Varian, 2000, p. 3). The actual history of e-commerce suggests many aspects of Varian’s predictions have been accurate. For example, the period from 1995 to 2000 was characterized by much experimentation, invention and capitalization. The lifting of the prohibition of commercial enterprises on the Internet, as referred to above, opened the doors for e-commerce to develop on a massive scale. The first web browsers were launched, and several companies were founded which are well known today as highly successful e-commerce enterprises, including Amazon.com, eBay and Alibaba, in China. Furthermore, Laudon and Traver (2016) describe a prevailing utopian vision emphasizing concepts such as perfect markets, disintermediation, selfgovernance, and “friction-free commerce” (p. 69). Such exalted confidence in a new technology and the associated economic paradigm is common for a great surge of development (Perez, 2002, 2009a, 2009b) and often accompanies a frenzy of investment and the consequent formation of a bubble (Orlikowski & Iacono, 2000; Perez, 2009a, 2009b). The end of this period corresponds with the beginning of the dotcom crash in 2000. The crash lasted from March 2000 to October 2002, and saw the Nasdaq Composite lose 78% of its value (Beattie, n.d.). Building on the work of economists such as Galbraith and Schumpeter, Perez interprets the dotcom crash as a cyclical event referred to as a major technology bubble (MTB) which are “caused by the way the market economy absorbs successive technological revolutions” (2009, p. 780). According to this view, MTBs regularly occur “midway along the process of assimilation of each technological revolution.” Four major precedents include Canal Mania during the Industrial Revolution, between 1771 and 1793 in England; Railway Mania during the Age of Steam and Railways occurring between 1829 and 1847 in Great Britain; multiple bubbles

17

related to the Age of Steel, Electricity and Heavy Engineering between 1875 and 1893 in the UK, USA, Argentina, and Australia; and the stock market crash of 1929 during the Age of Oil, the Automobile and Mass Production, which started in 1908 (Perez, 2009a, p. 9; Perez, 2009b, p. 782). Each of the four precedents is associated with great surges of development. Perez situates the dotcom crash within a fifth surge of development described as the Age of Information and Telecommunications beginning in 1971 in the USA (2009b, p. 782). As though on a fast-forward cycle, the dotcom crash seems to have propelled ecommerce from its experimentation and capitalization stages almost directly into a hypercompetition and consolidation stages during which many companies disappeared or were acquired. In reality, there is considerable overlap between Varian’s theoretical stages; many companies doubtlessly survived the crash by focusing on management, which is to say on operations and brand building. At any rate, after recovering rapidly from the dotcom crash, ecommerce grew by more than 10% a year from 2001 to 2006 and expanded beyond retail into more complex sectors such as financial services and travel. Laudon and Traver (2016) refer to this period broadly as one of consolidation. Significant events occurring during this period include both Alibaba.com and Amazon.com posting their first yearly profits, the acquisition of PayPal by eBay, and the creation of DHgate.com, China’s first business-to-business platform (DHgate.com, n.d.) Beginning in 2007, the cycle starts over to a certain extent even while overlapping with the trend towards consolidation. New technological developments allowing for greater interactivity and user-generated content on the Internet—developments which are often referred to as Web 2.0—and mobile phone technology have spurred another wave of experimentation and capitalization leading up to the present day (Laudon & Traver, 2016); meanwhile, Amazon.com

18

is estimated to have accounted for more than half of all e-commerce growth in 2015 (Garcia, 2015). This period has seen the meteoric development of the social networking site, Facebook, from a start-up company to a global phenomenon with an estimated 1.79 billion monthly active users today (“Company Info”, n.d.). Other companies which have come to prominence during this period, often thanks to venture capital, include the transportation network, Uber; the mobile photos sharing/video sharing social networking site, Instagram; the social networking site, Twitter; and the micro-blogging site Tumblr. Starting in 2008, mobile applications have become popular, and while the most well-known applications today are products of large, established companies, such as Facebook Messenger, Google Play, or Yahoo Stocks, many have been created by entrepreneurs, such as the photo-sharing site/mobile application Pinterest (Nusca, 2015). The cycle of experimentation and capitalization in e-commerce shows no sign of letting up. There are numerous recent trends and technological innovations still in their early phases of development. Some of these systems and technologies will most likely impact the future of ecommerce in important ways. For example, the first cryptocurrency, Bitcoin, a digital asset and payment system, was introduced in 2008, and has already been adopted by a significant number of billion dollar enterprises including Dell, Paypal, and Microsoft; the cryptocurrency relies on the Block Chain value-exchange protocol which holds enormous potential for the development of online transactions by facilitating micropayments, online tipping, and the registering and transferring of digital assets (Bheemaiah, 2015; Williams, 2015). Furthermore, artificial intelligence is now entering the business mainstream; Hardy (2016) cites several large and wellknown companies who have developed artificial intelligence products that are currently coming to market including Salesforce, a product which conducts real time analysis of data, and provides

19

“insights into what sales leads to follow and what products to make next” (Hardy, 2016, para. 2 and 3). Hardy (2016) also states that while enterprises and investors do not clearly understand what the business value of artificial intelligence is, they are aggressively investing in it and are optimistic about its future promise. Another frontier of experimentation, and one that has attracted staggering sums of capital investment, is a group of related technologies referred to as virtual reality, augmented reality, and mixed reality. One such company, Oculist Rift, a virtual reality headset maker founded by a 20-year-old entrepreneur in 2012, was acquired by Facebook in 2014 for $2 billion. Facebook alone has over 400 people dedicated to working on virtual reality systems (Kelly, 2016). Another company, Magic Leap, which creates mixed reality technology (described as “virtual reality overlaid on the real world”), recently raised what may be the largest amount of venture funding in history: $793.5 (Kelly, 2016, para. 4). It is not clear at this time how these new technologies will develop into business models but the implications are profound. Rony Abovitz, the founder of Magic Leap, considers that his company is “building the internet of presence and experience” (Kelly, 2016, para. 16). While the technological developments referred to as Web 2.0 vastly changed the e-commerce experience by opening up the possibilities for user-generated content and sophisticated interactivity, these new technologies may transform the e-commerce experience even more over the next decade. E-commerce retail context. The primary research conducted for this dissertation focuses on a French e-commerce company which sells its e-store expertise to apparel companies in France; these apparel companies contract the study company to manage their e-commerce activities. The relevant sector for this study is therefore the French e-commerce retail apparel sector. The services offered by the company currently range from a specific services offer, such

20

as the setting up and maintaining of e-commerce websites, to a wider package of e-commerce expertise which might include not only the setting up and maintaining of a site, but also after sales service, supply chain management, sales and promotion support, and strategic expertise. The staff of the company has regular contact with both the final consumer and the brand; the activity of this company can therefore be considered both B2B e-commerce, and B2C ecommerce. In the following section, I provide an overview of worldwide retail e-commerce, the context in Western Europe; I then narrow the focus to the context in France, and, where appropriate, provide specific figures about the apparel sector. Worldwide, the prospects for robust growth in the retail e-commerce sector are encouraging and Western Europe is an attractive region for e-commerce. In absolute numbers, Western Europe represents one of the highest concentrations of potential digital buyers and, moreover, represents a significant percentage of total retail sales. Worldwide retail sales represented approximately $22 trillion in 2015 with e-commerce retail representing approximately 7.4% of that total ($1.7 trillion), which is an increase of 5.6% compared to 2014; this share is predicted to further increase to 12.8% ($3.6 trillion) of worldwide retail sales by 2019 (Iankelevich, 2015). Despite that predicted increase, the share appears to be a relatively low percentage of overall retail, which suggests there may be room for more robust growth in the relative size of this share depending on factors such as Internet penetration, the development of online payment systems, and the percentages of Internet users in each country who conduct transactions online. Regarding the apparel sector, in 2014, clothing and clothing accessories (including footwear) was the largest retail merchandise category in the United States for ecommerce sales in the electronic shopping and mail-order House industry; it was valued at $46.8 billion which represents a 16.3% increase compared to 2013 (US Census Bureau, 2016, p. 2).

21

Furthermore, apparel is purchased by more than half of all online buyers in the EU and is the leading product category in several countries (YStats.com, 2016). Western Europe ranks as the second region worldwide after Asia-Pacific for the percentage of total e-commerce retail sales, and for the number of digital buyers; furthermore it is ranked number two after North America for digital buyer penetration (Iankelevich, 2015, p. 5). Western Europe is estimated to represent 8.2% of the total retail e-commerce sales compared to 12.4% for Asia Pacific; it is estimated to have 205.6 million digital buyers in 2016, which is 12.8% of the world’s population of digital buyers, compared to 921.6 million buyers in AsiaPacific, which is 50.4% of the world’s population of digital buyers. These Western European buyers represent 74% of all Western European Internet users estimated in 2016, and 57.6% of its total population, compared to 76.3% and 66.4% respectively for North America (Iankelevich, 2015, p. 5). Meanwhile, growth in the retail e-commerce sector appears to be slowing worldwide and is predicted to further decline. According to Eurostat (“E-commerce Statistics”, 2016), turnover generated by e-commerce in the EU-28 increased by 20% between 2012 and 2014; whereas, Iankelevich (2015) estimates growth in this sector at only 14.8% in 2014 in Western Europe, and predicts further declines from that point forward (p. 4). Part of the difference in estimates for the same year (2014) can be explained by the fact that the Eurostat EU-28 includes countries from Central and Eastern Europe which are not included in the Iankelevich definition of Western Europe. At any rate, according to Iankelevich (2015), Western Europe compares less favorably to other regions worldwide when it comes to predicted rates of growth in the e-commerce retail sector. For example, in 2016, Western Europe is ranked last after Asia-Pacific, Middle East & Africa, Central & Eastern Europe, Latin America, and North America; also, growth in Western

22

Europe is estimated at 10.5% in 2016, compared to 31.3% in Asia-Pacific, and is expected to slow to 7.3% by 2019 as compared to 23.5% in Asia-Pacific (Iankelevich, 2015, p. 4). Meanwhile, France compares strongly to other countries worldwide, and within Western Europe, both in terms of overall retail sales, and with respect to e-commerce retail. A global comparison of several countries places France, with $763.5 in predicted retail sales in 2016, number 6 after the United States, China, Japan, Germany, and India—all countries with substantially larger populations (Iankelevich, 2015). Within Western Europe, France is second in retail sales only to Germany which is predicted at $866.48 million, and ahead of the UK which is estimated at $701.70 (Iankelevich, 2015). Similarly, when it comes to retail e-commerce sales, France is also ranked 6 worldwide just behind Germany. Retail e-commerce sales in France in 2016 are predicted to represent $46.10 billion compared to $68.95 billion in Germany, which means that e-commerce retail sales in France will be 2.2% of total e-commerce sales worldwide compared to 3.4% for Germany in the same year—the biggest players here are China, predicted to account for 44.4% of all retail sales worldwide in 2016, followed by the US at 18.8%; also, with an estimated 30.7 million digital buyers in 2016, France is ranked 9th worldwide representing about 2% of the total number of buyers (Iankelevich, 2015). An analysis of retail e-commerce figures provided by Eurostat (E-commerce Statistics, 2016), however, suggest that several other countries in the EU-28 plus Norway are actually more competitive than France for the percentage of total retail represented by e-commerce turnover. At just over 20%, France is ranked 11th after Ireland (at 37%), Norway, Sweden, Denmark, Belgium, Czech Republic, Netherlands, the UK, and Slovenia (E-commerce Statistics, 2016). When the figures are compared according to the percentage of enterprises making e-purchases, France, at over 40%, moves up at least one position ahead of Slovenia, at less than 30%, and

23

possibly ahead of Denmark for which figures concerning this measure are not available (Ecommerce Statistics, 2016). Growth forecasts for France in e-commerce retail are somewhat contradictory. As stated above, the growth rate in e-commerce retail sales in Western Europe is predicted to lag behind other regions in the word. In France, Iankelevich (2015) reports that retail e-commerce sales grew by 11.1% in 2015, and predicts the rate to grow by only 8.2% in 2016 before slowing to a rate of 6.5% by 2019 (p. 15). Moreover, when it comes to expected rates of growth in the number of digital buyers, according to Iankelevich, France is predicted to fall; in 2014, France was ranked 11th worldwide with a growth rate of 5% for digital buyers (ahead of the United States) and is estimated at 2.6% in 2016, slipping behind countries such as the United States, Canada and Norway which move up in the rankings (p. 18). However, the French e-commerce federation (Fevad), estimates that growth in e-commerce turnover was 14% in 2015 (La Fédération du E-Commerce et de la Vente à Distance, 2015), which is significantly higher than Iankelevich’s estimates. Indeed, this estimate exceeded Fevad’s own predicted rate of 10%; moreover, the federation reports a robust growth rate of 16% in the first quarter of 2016 (La Fédération du E-Commerce et de la Vente à Distance, 2015). The federation also states that sales transactions made on smartphones and tablets progressed by 39% in 2015, and may be responsible for as much as 20% of all e-commerce sales in France (La Fédération du ECommerce et de la Vente à Distance, 2015). Reporting on consumer behavior, the French e-commerce federation states that while the average amount spent per e-commerce transaction in France declined somewhat—4.4% in the first quarter of 2016—the number of purchases online per person has increased by 13% with the annual overall average amount spent per consumer on e-commerce transactions increasing by

24

8%; while tourism represents the largest e-commerce sector in France in terms of turnover (32%) followed by apparel (10%), apparel leads tourism in terms of the number of products purchased (La Fédération du E-Commerce et de la Vente à Distance, 2015). Amazon is the leader in the ecommerce sector in France—it is visited by over 18 million Internet users each month, which represents 37.9% of the entire population of Internet users in France (Mediatrie, 2016); moreover, Amazon is moving aggressively into the apparel sector in the US and may surpass Macy’s as the number one clothing retailer by 2017 (Wahba, 2015), which suggests Amazon will also compete aggressively in this sector in France and elsewhere. To conclude, E-commerce appears to be a relatively small percentage of worldwide retail sales and enjoys a growth rate three times greater than brick-and-mortar retail (Iankelevich, 2015), and it will likely continue to grow faster than brick-and-mortar retail to take an ever increasing percentage of overall retail sales. While Western Europe is a key region when it comes to e-commerce as a percentage of total retail sales worldwide, and in terms of Internet penetration and digital buyer penetration, Western Europe lags behind Asia-Pacific and the Middle East in expected growth rates over the next five years. France compares well to other countries in the world in terms of overall e-commerce sales, number of digital buyers, and digital buyer population, even if it lags behind some EU-28 nations when it comes to a percentage of total retail turnover represented by e-commerce. Real growth rates for e-commerce in France, according to some estimates, are outpacing predictions but will most likely still not rival growth rates in countries in the Asia-Pacific or the Middle East. Apparel is one of the strongest sectors in France as it is in EU and in the US. But with overall e-commerce growth rates slowing and increased competition from players like Amazon.com, the e-commerce retail in the apparel sector is becoming more competitive.

25

Competitive advantage in e-commerce enterprises. As previously suggested, ecommerce should be seen in the context of the age of information and digital communications (Perez, 2009b), which might also may be referred to as the information economy (Shapiro & Varian, 1999). Hence, there is an obvious and necessary emphasis on technology and infrastructure when discussing e-commerce. An e-commerce company cannot exist without substantial use of and expertise in technology. Every employee of the study company must be well versed in the Internet, the World Wide Web, not to mention the software and/or platforms relevant to their specific activity. Every employee involved in the core operations of the company must possess technological expertise in one or more of the following areas: ecommerce platforms and related software, logistics and transportation platforms, project management software, data analytics, and/or web design, development, and integration—the list is not exhaustive. Technical innovations have been and will continue to be integral to ecommerce. Nevertheless, the role of technology should not be over emphasized. Technology, alone, does not determine the success or lack of success of the company, the development of the sector, nor the behavior in the workplace—although it will certainly have an impact. Ultimately, it is how people and businesses engage in technology in practice which gives technology its meaning and its value. Orlikowski and Iacono (2001) state that the digital economy is “an ongoing social product, shaped and produced by humans and organizations, and having both intended and unintended consequences” (2001, p. 357). They further describe technology as “social, dynamic, and multiple” (Orlikowski & Iacono, 2001, p. 358). Therefore, people, processes, and practices cannot be dissociated from the technology nor can their role be diminished.

26

Varian (2000) argues that in the Internet age, because companies can easily get information about each other’s products, services, and prices, competitive advantage will lie elsewhere. Moreover, as already noted in the Purpose of the Study, Christensen (1997/2011) contends that innovation is more about an organization’s processes and values than its resources; he suggests that even if you gave two different companies exactly the same resources, the results would be different. This is because the company’s specific processes and values are what allows it to “transform inputs into goods and services of greater value” (p. 187). Christensen (1997/2011) goes on to explain that processes may be formal, such as explicit and documented procedures; informal, such as routines which have developed over time; and, finally, that they may be cultural—which means ways of working and interacting that “have proven so effective for so long that people unconsciously follow them” (p.187). Furthermore he states that the values of an organization will determine what kind of projects the company will take on, and where and how it will invest its resources. Competitive advantage in e-commerce companies will therefore most likely not be found in products, services or technologies, alone, but, to a significant extent, in the company’s values, processes, routines, and in how the stakeholders (employees, business clients, and final consumers) engage in the technologies. Because, tacit knowledge tends to be embedded precisely in these areas, numerous scholars recognize its importance as a potential firm-specific strategic asset (Grant, 2013; Lazar & Robu, 2015; Nonaka, 1994; Spender, 1993, 2014). However, it would be a mistake to assume that the processes, routines, values, and practices are necessarily assets. They are only assets to the extent that they contribute to the company’s ability to create value. As Christensen (1997/2011) suggests, a company’s values may actually cause a company make poor investment decisions instead of good ones; likewise, processes are a

27

double-edged sword because they not only define an organization’s abilities but also its “disabilities” or limits because they are difficult to change once they are embedded in the organizations. This question is particularly important for companies moving out of the start-up stages and into a more established mode of operations. According to Christensen (1997/2011), over time the capabilities of the company will migrate from resources, such as key people, toward “visible, conscious processes and values, and then toward culture” (p. 195). When processes, values, and culture coalesce, as already stated above, they are difficult to change. It is vital that a company experiencing a migration toward processes and values, which are inseparable from tacit knowledge, be aware of what those processes are. Companies need this awareness to better understand what its capabilities are versus potential “disabilities”. Strategically speaking, the organization’s processes and values must be matched to the problem, set of problems or market opportunity at hand. Organizational Learning, Learning Organization, and Knowledge Management It is widely accepted that companies who wish to develop or maintain a sustainable competitive advantage must become aware of the tacit knowledge they possess and the extent to which it can be leveraged as a vital firm-specific resource (Grant, 2013; Nonaka, 1991; Prahalad & Hamel, 1990; Spender, 1993). Because it is often rooted in practice, and highly contextual, tacit knowledge is difficult to communicate; some researchers therefore emphasize the importance of transferring tacit knowledge to explicit knowledge in order to develop its strategic potential (Argote & Ingram, 2000; Barney, Wright, & Ketchen, 2001; Krogh, Nonaka, & Aben, 2001). Other researchers emphasize the importance of creating new knowledge from existing tacit knowledge (Bettis & Hitt, 1995; Bhat & Kumar, 2011; Cook & Brown, 1999; Nonaka,

28

1994, 1991/2007). Still others insist that a focus on practice is key to understanding how enterprises can best utilize, develop and leverage firm-specific knowledge and capabilities (Caldwell, 2012; McIver et al., 2013; Zander & Kogut, 1995) or how knowledge is embedded in routines and how developing new routines is related to developing new dynamic capacities (Andreu & Ciborra, 1996; Argyris & Schon, 1996; Ciborra & Andreu, 2001; Zollo & Winter, 2002). Finally, there are numerous works which focus on the organizational learning environment and how this is related to knowledge creation (Aggestam, 2006; Marquardt, 1996; Watkins & Marsick, 1993) and how to create organizations where knowledge creation will occur (Bui & Baruch, 2010; Goh, 1998; Senge, 1990/2006; Yang, Watkins, & Marsick, 2004). What companies know, how they know what they know, what they do with their firm-specific knowledge, how they encourage knowledge creation, and how they develop and leverage this knowledge are all key questions for this study. The following three sections of the literature review will focus on three fields in management and social sciences literature which contribute to our understanding of organizational knowledge and knowledge creation, and which are relevant for this dissertation: organizational learning, learning organization, and knowledge management. Organizational learning. Organizational learning developed out of the disciplines of behavioral psychology and organizational behavior in the late 1970s—it is concerned with the creation, retention and transfer of knowledge within an organization as well as learning processes and outcomes (Argote, 2011; Argote & Miron-Spektor, 2009; Argyris & Schon, 1996; Dodgson, 1993). Gunasekara (2003) notes the difficulty of finding a clearly defined and practical understanding of the term, and also states that it may be used interchangeably with the term “learning organization” (p. 37). He refers to two definitions, each of which makes use of the

29

word “process”. One is Matthews’ (1999) definition which states that “workplace learning involves the process of reasoned learning towards desirable outcomes for the individual and the organization” (p. 19-20, as cited in Gunasekara, 2003, p. 38). The other is Fiol and Lyles’ (1985) definition which maintains that “organizational learning means the process of improving actions through better knowledge and understanding” (p. 803, as cited in Gunasekara, p. 38). However, a more recent definition from Argote and Miron-Spektor (2009) emphasizes change from acquired experience rather than processes: “organizational learning is a change in the organization that occurs as the organization acquires experience” (p. 4). One of the most often referred to and influential concepts found in the organizational learning literature is “double-loop learning”. According to Argyris (2002), “double-loop learning occurs when errors are corrected by changing the governing values and then the actions” (p. 206). He contrasts this to “single-loop learning” which is when errors are corrected “without altering the underlying governing values”. To illustrate the difference, Argyris describes a thermostat that is programmed to turn on if the temperature in a room becomes too cold and to turn off if it becomes too hot. He contrasts this to the example of a thermostat that “questions why it is programmed to measure temperature, and then adjusts the temperature itself” (p. 206). The first thermostat illustrates “single-loop learning” (i.e. correcting errors), while the second illustrates “double-loop learning” (questioning/changing underlying values). Heorhiadi, La Venture, and Conbere (2014) refer to the former as “informative” learning which “allows people to learn more about the thing that fits their mental models” and to the latter as “transformative learning” which is the “process of changing mental models” (p.7). Jensen (2005) also insists on the crucial role of “double-loop learning” in the process of transforming data into information and the process of transforming information into new knowledge.

30

Whether organizational learning deals with one entire organization or subgroups within the organization, the learning construct must be group and not just individual; an aggregation of what individuals learn will not be sufficient to be considered group learning. Wilson, Goodman, and Cronin (2007), building on a number of prior definitions of group learning and scholarly work on the construct, including that of Edmondson (1999, 2002), Argote, Gruenfeld, and Naquin (2001), Ellis, Hollenbeck, Ilgen, Porter, and West (2003), Gibson and Vermeulen (2003), Gruenfeld, Martorana, and Fan (2000), and London, Polzer, and Omoregie (2005), state that “group learning is essentially about the internal and external manifestations of information processing” (p.1044). These authors propose four criteria to be considered in a theory of group learning. First, the level of analysis must be the group, as already noted above. Referring to the work of Klein, Dansereau, and Hall (1994), Wilson et al. (2007) argue that “the theory, measurement, and analysis of group learning should focus on changes in the group’s repertoire” (p. 1043). Second, the construct of group learning must include the fundamental processes of sharing, storage and retrieval of group knowledge, routines, or behavior. Third, referring to the work of Huber (1991), they define learning as an outcome, which is to say “a change in the range of the group’s potential behavior” (p. 1044)—distinguishing this from performance. Lastly, in order to distinguish the concepts from other phenomena like collective decision making or problem solving, their conceptualization of group learning requires that the change in the group’s repertoire take place over a time interval. To illustrate the complex conceptualization of group learning developed by Wilson et al. (2007), we can imagine a work team which holds a meeting to talk about a recurrent problem. The members share different pieces of information and different points of view about it. This knowledge about the problem is discussed in order for the group to make sense of it collectively.

31

The group then decides on a new approach to the problem which is stored in each individual’s memory and recorded in minutes. The next time the problem occurs, the members retrieve the knowledge of their newly developed approach from their respective memories and from the meeting minutes. Their collective approach to the problem is effective and the problem is dealt with better than before. This becomes a new model for the group to deal with this recurrent problem. In this manner, four criteria described above are met (group level; fundamental processes; an outcome; and the change occurring over time). Learning organization. The concept of a learning organization developed primarily in the 1990s although interest in the subject can be traced to the 1960s (Kirwan, 2013; Senge, 1990/2006; Watkins & Marsick, 1993; Yang et al., 2004). While many scholars have made contributions to the development of the learning organization concept, including Pedler, Burgoyne, and Boydell (1991), Garratt (1995), Watkins and Marsick (1993), and Marquardt (1996), it is the work of Senge (1990/2006) which is most often referred to. According to Senge (1990/2006), learning organizations are those “where people continually expand their capacity to create the results they truly desire, where new and expansive patterns of thinking are nurtured, where collective aspiration is set free, and where people are continually learning how to learn together” (p. 8). Senge proposes five disciplines that are key in building learning organizations: systems thinking, personal mastery, mental models, shared vision, and team learning. His learning organization concept is complex and builds on the work of others, as he frequently acknowledges. For example, his notions of organizational learning owe much to the work of Argyris and Schon (1996). Likewise, the discipline of “mental models” comes from the work of Johnson-Laird (1996), and systems thinking from Forrester (2007). It is this latter discipline,

32

systems thinking, which is most associated with Senge’s version of the learning organization concept. Garvin (1993/2000) characterizes Senge’s description as both “idyllic” and “desirable” and suggests the need for an actionable framework. In attempt to build such a framework, Garvin proposes a definition of the learning organization which he intends to be actionable, to provide guidelines for practice, and to be measurable: “A learning organization is an organization skilled at creating, acquiring, and transferring knowledge, and at modifying its behavior to reflect new knowledge and insights” (Garvin, 1993/2000, p.80). He goes on to assert that learning organizations must be skilled in five main activities: “systematic problem solving, experimentation with new approaches, learning from their own experience and past history, learning from the experiences and best practices of others, and transferring knowledge quickly and efficiently throughout the organization” (Garvin, 1993/2000, p. 81). He furthermore recommends steps that companies may take to build learning organizations, including: fostering an environment conducive to learning, opening up boundaries and stimulating the exchange of ideas, and creating learning forums (p. 91). Indeed, much of the recent scholarly work done on the learning organization concept has been directed at rendering the concept empirically measurable and applicable (An & Reigeluth, 2005; Bui & Baruch, 2010; Garvin et al., 2008; Heorhiadi et al., 2014; Jagasia, Baul, & Mallik, 2015; Kirwan, 2013; Lazar & Robu, 2015; Shipton, Zhou, & Mooi, 2013; Yang et al., 2004). According to Argyris and Schon (1996), the learning organization is actually a branch of organizational learning literature—one which is promulgated mainly by consultants and practitioners; the authors contrast this branch with the “skeptical” organizational learning literature produced mainly by academics (p. 180). Even while conceding that learning

33

organization literature has made a valuable contribution to the field of organizational learning, Argyris and Schon (1996) argue it does not generally give enough importance to theoretical questions, and rests on certain problematic assumptions—such as the beneficence of organizational learning—without giving adequate consideration to questions of validity and utility (p. 187). Caldwell (2012) is also particularly critical of Senge’s systems thinking version of the learning organization concept; he states that “Senge’s learning organization tends to give primacy to structure over agency, system over action, consensus over conflict, norms over practices” and concludes that it should be abandoned as a “theoretical and practical guide to organizational change” (p. 159). It appears that the learning organization concept has failed to achieve a solid consensus in the scholarly literature and, as a field, it is fragmented. Despite this weakness, the concept, and the field of literature still have value for this study. The learning organization should be considered an ideal type, or archetype of an organization which excels in certain organizational learning practices and achieves outcomes. A number of scholars support this position, including Rowden (2001) who considers the learning organization as a model for organizational change, and King (2001) who states that is a “goal to be pursued rather than a state of affairs to be achieved” (p. 14), and who proposes a number of different strategies for pursuing that goal. While it may be less academically sound compared to what Argyris and Schon (1996) consider to be the more scholarly branch of learning organization literature, the learning organization literature contains much that can be used by companies and managers—notably best practices and tools for evaluating the effectiveness of those practices, and for assessing the perception and attitudes of employees (Edmonson, 2008; Garvin et al., 2008; Lazar & Robu, 2015; Senge,

34

1990/2006). Furthermore, recent scholarly work such as Jagasia et al. (2015) and Shipton et al. (2013) demonstrate that the concept is still receiving serious scholarly attention. Knowledge management. Knowledge management, a discipline which has been evolving since the 1990s, is concerned with many of the same issues as organizational learning. McIver et al. (2013), citing Foss and Mahnke (2003) state that “[knowledge management], a set of management activities aimed at designing and influencing knowledge creation and integration (including processes of sharing knowledge), has emerged as a particularly influential organizational competence that shapes the work environment” (p. 597). Writing about the origins of this discipline, Prusak (2001) distinguishes it from organizational learning by arguing that knowledge management recognizes the importance of questions overlooked in the former discipline, such as how learning occurs, and what kind of business outcomes can be expected from it. He also states that knowledge management focuses on “value as a function of user satisfaction” (p. 1005) putting the emphasis on concrete, situation-specific value as opposed to more theoretical and abstract approach. Both Argote (2005) and Spender (2014) assert, however, that the field has evolved in two distinct streams. Argote (2005) describes, for example, two markedly different paradigms found in the knowledge management literature: The computational view of knowledge management approaches knowledge as identifying empirically validated facts and managing them through technology. By contrast, the organic paradigm includes the people, group dynamics, social network, and cultural aspects of knowledge management and includes tacit as well as explicit knowledge. Thus, the organic paradigm is more encompassing than the computational. (Argote, 2005, p. 45) Spender (2014) likewise refers to two paradigms in the field: one positivist (similar to Argote’s “computational” paradigm) the other subjectivist (similar to the “organic” one). He states that the knower-centered subjectivist paradigm deals with knowledge “as an interplay of discrete and experienced modes, none comprehensive, all partial” and contrasts this to the

35

positivist paradigm which “loses the knowing person” in its rational and abstract assumptions about how things work (p. 7). He emphasizes the subjectivist paradigm and deplores that most writers in the field identify more with the positivist one. Moreover, he goes beyond a position of opposing paradigms and proposes a third paradigm referred to as “experimental practice”; he argues that “experimental practice binds the subjective and objective paradigms in potentially useful ways” (p. 8); he further states that this third paradigm allows for a new definition of the firm as “an integrated community of skillful practices” (p. 8), and contrasts this to more mechanistic theories of the firm which focus primarily on inputs and outputs. It is in the knowledge management literature where different types of knowledge (such as tacit and explicit), as well as knowledge creation are most clearly addressed. Spender (1993), for example, contends that “competitive advantage arises from the interaction of many different types of knowledge” (p. 39). He describes four such types, two of which are individual (conscious, and automatic) and two of which are social (scientific and communal). While scientific knowledge is explicit, the other three (conscious, communal and automatic) are implicit, in his view. Writing about the importance of context, Spender refers to the work of Brown and Duguid (1991) on copy machine repairmen which demonstrated that the explicit (scientific) knowledge of how the machines work was insufficient to resolve problems because the problems often occurred “in the relationship between the customer and the copier” (p. 39). Repairmen had thus to situate their knowledge in the context of the user’s practice. Furthermore, an important pool of knowledge for resolving problems was communal—when the repairmen would talk to each other informally, proposing solutions and working towards a “common diagnosis” (p. 39), hence creating what is often referred to in the literature as a “community of practice”. To further illustrate his point, Spender describes the example of EMI’s failure to

36

leverage their own invention, the CAT scan, into a sustainable competitive advantage. According to Spender, while the company had a powerful patent (hence scientific knowledge), they lacked a number of other crucial aspects of knowledge such as tacit knowledge of how the medical imaging business worked, how the communities of practice worked (radiologists, and suppliers), not to mention marketing and manufacturing knowledge. Nonaka (1994), meanwhile, describes four modes of knowledge creation: socialization, combination, externalization, and internalization. Socialization is when tacit knowledge rooted in experience is shared through observation, imitation, and practice. Combination is when social processes such as meetings and conversations allow for existing explicit knowledge to be reconfigured in a new way. Externalization is when tacit knowledge becomes explicit; and internalization is when explicit knowledge becomes tacit through assimilation. According to Nonaka (1994), “organizational knowledge creation, as distinct from individual knowledge creation, takes place when all four modes of knowledge creation are ‘organizationally’ managed to form a continual cycle” (p. 20). He describes this cycle as a “spiral” and illustrates this with the example of building a team in which members share experiences and perspectives (socialization), which allows for dialogue in which members “reveal hidden tacit knowledge” that is otherwise difficult to communicate (externalization); this is followed by the team generating concepts and ideas which can then be combined with existing data and knowledge (combination); and then assimilated through experimentation and practice (internalization) (p. 20). It is interesting to note here some similarities between Nonaka’s (1994) conceptualization of organizational knowledge creation and the conceptualization of group learning developed by Wilson et al. (2007) described above. Similarities can also be drawn to the organizational learning framework developed by Crossan, Lane, and White (1999) which

37

includes the four key processes of intuiting, interpreting, integrating, and institutionalizing. These similarities are discussed in more depth in Chapter 5, in the Theory subsection. The work of Cook and Brown (1999), which builds on the work done in the field of organizational learning and knowledge management, attempts to refine our epistemological understanding of different types of knowledge and to challenge the assumption that it is possible to transfer one type of knowledge to another. They maintain that that four types of knowledge (individual explicit; group explicit; individual tacit; group tacit) are distinct forms of knowledge on equal footing. By taking this position, they differ from most of the literature in the field of knowledge management, which uses terms such as “transfer” and “conversion” (for example, Nonaka). Cook and Brown (1999) contend that individual tacit knowledge cannot be “converted” or “transferred” into explicit knowledge. Moreover, they propose that while “knowledge” is static, “knowing is dynamic, concrete, and relational” (p. 387), and suggest that the epistemology of “knowing”, closely linked to practice, be given as much importance as knowledge. Here we find an echo with Spender’s argument, cited earlier that “experimental practice” forms the basis of a “third paradigm” in knowledge management literature (2014). Cook and Brown define practice as “the coordinated activities of individuals and groups in doing their ‘real work’ as it is informed by a particular organizational or group context” (p. 386-387). The importance of their work to this study lies in the understanding that practice should be considered part of the dynamic knowledge cycle, and that there is an inherently creative element to developing one type of knowledge (explicit) based on another type of knowledge (tacit). How the fields are interrelated. The three fields have many concerns in common— establishing clear boundaries between the fields is indeed more challenging than identifying areas where they overlap. Aggestam (2006), for example, maps the learning organization—

38

based primarily on Senge’s version of the concept—to knowledge management using the following key words: culture, leadership/management, vision, work processes, organizational learning, external factors, internal factors, system’s thinking, organizational memory, and technical (p. 297). He concludes that “external factors” and “system’s thinking” are unique to the learning organization while “organizational memory” and “technical” are specific to knowledge management. According to Aggestam, the two concepts are on “different levels of abstraction with different focus and purpose” (p. 298); his conceptual model places knowledge management as a subsystem of the larger system which is the learning organization described as “an environment that promotes a culture of learning” (p. 301). Argote (2005), in a commentary on two essays, one dealing with organizational learning, and the other with knowledge management, explores the relationship between these two fields. She states that “organizational learning focuses more on the process through which organizations acquire knowledge from experience, whereas knowledge management focuses more on managing what is learned” (p. 43). Much of her essay, however, is devoted to exploring common ground. For example, she insists on the multidisciplinary nature of both fields; the real problems they aim to address involve aspects of psychology, sociology, operations management, organizational behavior, strategic management, economics, and information systems (p. 45). While describing this multidisciplinary nature as a strength she also maintains it can function as a weakness because it makes the “culmination of findings” difficult as researchers often mean different things when using the same terms. While she does not argue for a convergence of approaches, she does propose that researchers and academics “cumulate findings within each approach so that our understanding of the phenomena advances” (p. 46). Furthermore, she suggests a few examples of areas which have seen successful “accumulation” between the two

39

fields such as transactive memory, knowledge transfer across organizational units, the impact of personnel rotation on knowledge transfer, and the relationship between social networks and knowledge transfer. In sum, she proposes a synergistic relationship between the two fields and suggests directions that future research could take: “if we had a deeper understanding of the processes through which organizations learn, we could design better knowledge management systems to capture and transfer knowledge acquired through that learning” (p. 46). Theoretical Framework for the Study This study will draw primarily on knowledge management literature and theory as the foundation for its research approach, and will draw to a lesser extent on the fields of organizational learning and the learning organization. Knowledge transfer is an area of interest to both organizational learning and knowledge management (Argote, 2005). Spender’s, (1993) different types of knowledge, and emphasis on context as well as practice, Nonaka’s (1994) different modes of knowledge creation, and Cook and Brown’s (1999) focus on the epistemology of knowledge and knowing provide theoretical justification to this study because they stress the importance of tacit firm-specific knowledge, how this knowledge should be conceived of as part of a dynamic knowledge cycle, and how knowing and practice are inseparable. Furthermore, the work of Ambrosini and Bowman (2001) will provide a basis for operationalizing the concept of tacit learning. Citing the work of numerous scholars, including Nonaka, Polanyi, Ravetz, Sternberg, Spender, Zander, and Kogut, these authors argue that tacit knowledge is difficult to formalize, tends to be personal, is embedded in practice, and is context specific. In order to address the inherent challenge of studying something which is not easily expressible, they posit the existence of degrees of tacitness; while “deeply ingrained tacit skills” might be inaccessible, other tacit skills could be articulated—even if imperfectly—and therefore

40

captured or observed in a study. Claiming a social constructivist stance—they cite Berger and Luckmann (1966)—Ambrosini and Bowman propose a methodology to conduct empirical research on tacit skills. The methodology involves conducting focus groups or semi-structured interviews as a basis to create causal maps, which are defined as “a form of cognitive map that incorporates concepts tied together by causality relations” (Weick & Bougon, 1986, p. 106, as cited in Ambrosini & Bowman, 2001, p. 817). The focus groups or semi-structured interviews are used to elicit viewpoints that allow the researcher to represent subjective data, to establish context by placing concepts in relation to one another, to give order on a micro-level to material which is inherently disordered, and to show multiple consequences and interrelationships (Eden & Ackermann, 1998, as cited in Ambrosini & Bowman, 2001, p. 818). The authors suggest encouraging participants to tell stories and, like Nonaka (1994), insist on the power of metaphors as a technique for rendering the implicit explicit. McIver et al. (2013) propose a theoretical framework that will also be used in this study to build an empirical research tool to assess knowledge management activities in the company. Building on the work of Brown and Duguid (2001), Cook and Brown (1999), and Orlikowski (2002), McIver et al. focus their level of analysis on the “work setting”, which is to say on practice. Their theoretical framework, referred to as “knowledge-in-practice” is formed from the two dimensions of knowledge tacitness and learnability. Different work settings are classified according to the degree of tacitness of the practice, and according to the degree of learnability of the activity. For example, activities such as that of a grocery store cashier or a bank teller are characterized as being low in tacitness and high in learnability, whereas the activities of an artist or an inventor are characterized as being high in tacitness and low in learnability. The intent behind the theory is to align the knowledge-in-practice type with the most suitable knowledge

41

management tools to produce the most desirable outcomes. In a work setting, such as that of a grocery store cashier, the best types of knowledge management activities might be the creation of a how-to manual, and the development of standardized procedures in order to produce desired outcomes such as codified and available information, and efficient routines. On the other hand, in a work setting associated with an inventor in an organization, the best types of knowledge management activities might include developing expert yellow pages and encouraging participation at specialized conferences in order to produce desired outcomes such as sharing expertise and the development of new ideas. While the above mentioned works provide the theoretical foundations for conducting qualitative empirical research on tacit knowledge (or skills), and knowledge management activities, the work of Garvin et al. (2008) will provide an additional quantitative survey instrument relating to the learning organization. Garvin et al. (2008) state that there are three building blocks to learning organizations: “(1) a supportive learning environment, (2) concrete learning processes and practices, (3) leadership behavior that reinforces learning” (p. 110). Designed as a diagnostic tool to foster learning, their survey tool includes a set of questions to assess the strength of learning in the organization for each building block. The survey can be taken by individuals within the organization; scores are then calculated for each building block. The scores can be compared to benchmark scores established by the authors; and, furthermore, scores of different units within the organization can be compared to each other. In this study, the survey instrument will provide a way to objectively assess the learning environment and practices in the company, and to compare the environment and practices both internally and externally.

42

Chapter 3: Research Design and Methodology Purpose of the Study As previously stated, the purpose of the study is to assess the tacit knowledge possessed by employees of a young e-commerce firm to help identify critical organizational skills; to explore ways of generating new explicit knowledge from this tacit knowledge through practice so that the company may leverage its organizational skills and knowledge as sustainable competitive advantages; and to assess the depth of learning in the company with a view to identifying areas of strengths and weaknesses. This study will provide an example of how to assess an e-commerce company’s environment, processes and values; and it will have implications for leveraging the insights gained for strategic value. The research problems this study is intended to address are as follows: a)

How are the concepts of organizational learning, knowledge management, and the learning organization related?

b)

What are different types of organizational knowledge and how are they related?

c)

How can firm-specific tacit knowledge of employees in an organization be mapped to help identify crucial organizational skills?

d)

What types of practices in the company contribute to the creation of new knowledge, and how might the company leverage those practices to develop new explicit knowledge which can potentially be leveraged as a sustainable competitive advantage?

e)

How can learning be assessed in an organization with a view to identifying strengths and weaknesses?

43

Description of the Company and Population Being Studied The company featured in this study was founded in 2008 by three young professionals in their late twenties and/or early thirties who met while working at an e-commerce start-up firm in France. They decided to leave the company for which they worked to found their own ecommerce start-up firm. The three of them were all enthusiastic about the potential development of e-commerce and anticipated major changes in the apparel sector. One of them was particularly strong when it came to knowledge and experience in Internet-related technologies and the World Wide Web; another had experience and a strong interest in fashion, photography, and images; and the third founding member was well versed in business models and how to finance new ventures. Together, they created a full-service e-commerce company whose business-to-business offer specifically targeted apparel brands. Beyond simply providing a service, the founders’ vision was to accompany the brands in the development and implementation of their e-commerce strategies. The firm was able to offer not only the knowhow and technical expertise required for the design, development and operation of the brands’ ecommerce websites, but also a full range of services including logistics management, after sales service, search engine optimization, data analytics, and e-marketing strategies. The revenue model depended on a substantial sales commission of 40%. The company grew rapidly—only six years after its creation, the company employed 90 people, had 17 clients and occupied three floors of open space in a warehouse-like facility in a major French city. At the time of the study, eight years after its creation, the company employs 115 employees, enjoys an annual turnover growth rate of 30%, and has 30 clients in the apparel sector. They maintain 25 active e-commerce sites which receive a combined total of between 50 and 60 million visits per year. The company generates from 40 to 50 million euro in turnover

44

annually. One of the three founders left her position in the company in 2015, and a second cofounder has reduced his operational and development role to focus mainly on the photographic and image aspects of the websites. The remaining co-founder is currently president of the company. The three founders remain on good terms and own the majority of the privately held shares in the company. Consistent with the founders’ vision to minimize differences in status internally, the company includes only three levels of hierarchy: the president and the executive committee, the top managers, and the employees. Six members of the executive committee report to the president/co-founder; they represent respectively the following six functional areas: finance and human resources; sales and strategy; e-commerce; operations (web integration, IT development/project, and logistics); marketplace; and photography. Each member of the executive committee has from 12 to 20 years of professional experience. The next level of the hierarchy includes 11 top managers who report to specific members of the executive committee. Each manager has from five to 10 years of professional experience. The remaining employees are primarily technical employees and/or lower-level managers who have less than five years of professional experience and work as web developers, web designers, web integrators, e-store managers, business developers, digital growth managers, and/or who work in the various functional departments such as after sales service, finance, and human resources. In the fashion/apparel sector, there are approximately seven levels of quality and price range. They are presented below: 1)

Top luxury

2)

Luxury

3)

Entry luxury

45

4)

Premium market

5)

Mid-level market:

6)

Mass market

7)

Discount

The company deals primarily with brands representing the mid-level, premium and entry luxury markets. The premium market, which generally has a price range from €200 to €400 per item, has recently been experiencing the highest level of growth. The two largest company departments are e-commerce and operations. E-commerce comprises the functional areas of digital traffic analysis, after sales service, brand management, and web design; this department alone employs about 40 members of the staff—thus it represents nearly half the company. Operations comprises web integration, IT/development, projects, and logistics; this department employs about 20 members of staff. The other departments employ fewer than 10 members of staff. The smallest department, photography, includes only two people. Marketplace, a department employing from four to five people, was created only one year ago and will likely soon be integrated into the e-commerce department. The broad goal of marketplace, to help companies sell their products through online marketplaces (see the Chapter 2 subsection Definition of e-commerce), overlaps with the goal of the e-commerce department. But because marketplace was a new initiative, it was set up as a distinct department; the management of the company felt this would be the best way to launch this new activity. While the company has been very successful in gaining market share, has grown exponentially in its relatively short history, and is now one of the best known B2B e-commerce companies in its sector, it is not yet profitable. Therefore, the company is making significant changes in its business model. For the past year, when dealing with new and potential clients,

46

the company has been working to diversify its business model and to debundle its services so that the full-service e-commerce offer is proposed as one option among others. The current sales and strategy team, which had been in place for a little more than one year at the time of the study, takes a tailored approach to clients; they try to identify specific e-commerce needs and aspirations, and to satisfy only those specific needs. Commission is now on a sliding scale from 11% to 23% depending on volume. This approach has been adopted for two reasons. First, many B2B clients in the apparel sector now prefer a more targeted offer then was the case when the company first started because these clients are more knowledgeable about e-commerce and have more definite ideas about their e-commerce strategies. Second, the company has recognized that some services are more profitable than others. Moreover, the company is developing its own e-commerce platform which it intends to be more adapted to the apparel sector than existing e-commerce platforms, such as Magento. Their plan is to license the platform to clients who would then potentially have a large degree of autonomy in their ecommerce operations. As stated previously, the founders’ vision for the company from the beginning has been to help clients grow and develop their e-learning activities, sales, and strategies. Its added value to the apparel brands is its combination of e-commerce technical and marketing expertise in the sector. The company is in the process of redefining its offer so that it proposes a portfolio of services that clients can choose from rather than having to accept or refuse one package offer— services such as after sales services, logistics management, and photography are not considered their core areas of expertise, and are offered to new and potential clients as extra services priced according to a rubric. Furthermore, over the past year, the company has developed the marketplace department to help brands take advantage of multiple existing online marketplaces

47

as additional sales channels; the company has also created a digital media position whose role is to provide advice and expertise to clients about digital media communication channels such as social networking sites; finally, as already mentioned, the company is developing a new ecommerce platform which it hopes to license to clients. Currently, however, the majority of the clients are still on the full-service contracts. This redefining of the offer has happened at the same time that the company has made internal staff management changes. The chief financial officer position was created less than a year ago as well as the position of human resources manager. A human resources manager was hired shortly after the chief financial officer, and now has three people reporting to her. Before the creation of this position, there was very little formalization of human resources policies— employees had no interlocutors other than their managers. The new human resources manager has been given a budget of €200K to spend on training and staff development. The company’s vision to help their clients grow and develop is mirrored in their internal vision for staff management: to help staff grow and develop. Employees are therefore encouraged to be proactive and autonomous problem solvers at every level of the hierarchy. The current objective for top managers is to reduce the percentage of operational duties to less than 30% of their daily activities so that they can devote the large majority of their time to people management, and staff development. The company is currently testing an approach to training, which is inspired by interval training in sports, whereby which one employee will train another in short, highintensity sessions. The data was collected on representatives of the company at all three levels of the hierarchy. Individual interviews included three members of the executive committee: the cofounder and head of photography, the head of commercial development and strategy, the chief

48

financial officer, and the e-commerce manager. All four are men in their mid-to-late forties who have master’s degrees in management, finance, and/or business, and have from 12 to 20 years of professional experience. Two of them have experience working outside of France—the ecommerce manager has over 10 years of high-level experience in the e-commerce sector in the United States. The co-founder has obviously been with the company since its creation, the ecommerce manager had been with the company for a little over two years at the time of the study, while the chief financial officer and the head of commercial development and strategy had been with the company for a little more than one year. The three top managers who participated in the individual interviews include the human resources manager, a brand manager, and the after sales service manager. The human resources manager had joined the company one year prior to the study, the brand manager had joined the company a little over a year prior to the study, and the after sales service manager had been with the company for a year and a half. All three are in their late thirties to mid-forties and have from five to 10 years of professional experience. Two of them are female (the human resources manager, and the brand manager) and one is male (after sales service manager). The 34 employees who participated in the group interviews hold a wide range of positions such as web developer, business development manager, project manager, web designer, financial controller, e-store manager, recruitment manager, logistics and transport manager, traffic manager, and digital media manager. They have been with the company anywhere from three years to six months, and generally have fewer than five years of professional experience. All have post-secondary degrees in various fields such as web development, marketing, and fashion. Most of them are in their thirties. Fourteen of the employees who participated in the group interviews were male and 20 were female.

49

The enterprise was chosen for this study because it is an example of a young, innovative technology company whose profile matches the phenomenon described by Christensen (1997/2011), whereby the capabilities of the company are migrating from resources such as key people toward “visible, conscious processes and values” (p. 195). It therefore provides an excellent opportunity to assess tacit and explicit learning. The young ages of the founders and most of the employees, and the enterprise’s cutting edge expertise in a dynamic sector suggests the enterprise is representative of emerging trends in the work place. Also, the company size is suitable for an instrumental case study research design (Stake, 2005). Furthermore, the company’s management recognizes the importance of organizational learning and is interested in the concept of the learning organization—hence their willingness to participate in the study. The fact that a large percentage of company employees participated (36%), the diversity of their roles, and the different levels of the company hierarchy represented in the study suggest that the sample reliably and validly captures the attitudes and perceptions of the entire population. The participants willingly participated in the study after having been solicited by the management, namely the e-commerce manager and the human resources manager – therefore, the sample should be considered a convenience sample. Design Overview Research design. This theory-driven study focuses on one company and applies a mixed methods cross-sectional survey research approach using a quantitative survey tool, and a qualitative semi-structured interview protocol. The quantitative survey tool, described in more detail below, is the Learning Organization Survey (Garvin et al., 2008), which is based on the theory of the learning organization described in Chapter 2. The qualitative semi-structured interview protocol, described in more detail below, is based on the causal mapping methodology

50

developed by Ambrosini and Bowman (2001), and on the knowledge-in-practice framework developed by McIver et al. (2013). The overall research design fits the definition of a case study according to Creswell (2011), Stake (2005), and Yin (2013). According to Christensen, Johnson, and Turner (2011), survey research is defined as “a research method where individuals fill out a questionnaire or are interviewed about their attitudes, activities, opinions, and beliefs” (p. 330) and is widely used in non-experimental research. According to Pinsonneault and Kraemer (1993), “the purpose of survey research in description is to find out what situations, events, attitudes, or opinions are occurring in a population” (1993, p.80). In this type of research, the questionnaire and/or interview protocol are standardized so that each participant is presented with the same stimulus such as instructions and/or directions (Pinsonneault & Kraemer, 1993). Both the survey and the semi-structured interviews were done in-person which has a number of advantages. Citing Webster (1997), Evangelista, Poon, and Albaum (2012) argue that “compared to telephone or mail surveys, personal contact is much more flexible, allows for the collection of a greater variety of data, and is better when questionnaires are lengthy” (p. 1175). Pinsonneault and Kraemer (1993) likewise contend that the quality of face-to-face interviews is higher when compared to other methods such as mail questionnaires. As discussed more in the subsection Learning Organization Survey, I used the groupadministered questionnaire method to have the participants complete the quantitative survey before the semi-structured interviews; the survey was completed in the same fashion before each individual interview. I convened the participants in individual or group sessions, and handed out the survey(s) to the participant(s) who completed it during the sessions. Generally, this method, whose main advantages are its speed and efficiency, is used with “organizational surveys when

51

the participants are available in a single workplace” (Christensen et al., 2011, p. 338). The quantitative survey tool, the Learning Organization Survey (Garvin et al., 2008), measures the multidimensional construct of the learning organization using a Likert Scaling multichotomous response format, also known as a summated rating scale (Christensen et al., 2011). The survey uses closed-ended questions which require respondents to choose from “a set of predetermined response alternatives provided by the researcher” (Christensen et al., 2011, p. 343). The semi-structured interview protocol, on the other hand, incorporated open-ended questions which are frequently used in exploratory and qualitative research. In open-ended questions, participants respond “in their own natural language, and they are not limited to a set of predetermined categories” (Christensen et al., 2011, p. 343). The role of the interviewer is crucial in semi-structured interviews; Gendal (1999) asserts that “the approach taken to gain cooperation of the respondent such as being respectful, non-demanding or somewhat deferential, will make a significant difference” (as cited in Shaw, Bednall, & Hall, 2002, p. 535). The overall research design of the study fits the definition of case study research. Stake (2005) asserts that the object of a case study must be sufficiently specific and bounded to constitute a case; it must not be too vast for the researcher or a small team of researchers to be able to experientially learn from it; otherwise the researcher will be forced to rely too heavily on abstractions, and on instruments or impersonal reporting. Stake (2005) defines three types of case studies, all of which have different purposes: intrinsic case studies, instrumental case studies, and multiple case studies (also called collective case studies). An intrinsic case study is one in which the case is interesting by itself not because of its potential relationship to an abstract construct or category, and not because it illustrates a more general problem. An instrumental case study, on the other hand, is meant “to provide insight into an issue or to redraw

52

a generalization” (Stake, 2005, p. 445). Finally, the multiple case study is described as follows: “When there is even less interest in one particular case, a number of cases may be studied jointly in order to investigate a phenomenon, population, or general condition” (Stake, 2005, p. 445). Stake’s definition of an instrumental case applies to this study which aims provide insight into issues of organizational learning, the learning organization, and knowledge management within the specific, bounded context of a single company. Creswell (2011) defines the case study as a research design in which the “researcher explores in depth a program, an event, an activity, a process, of one or more individuals” (p. 15). For this study, the “individual” corresponds to the single company, which is the object of the research. In case study research, the in-depth focus on a single object of study affords the researcher the opportunity to gather data from multiple sources, and to analyze these data in different ways thus allowing for a more complete and richer understanding of the phenomenon of interest and its relationship to the context (Pinsonneault & Kramer, 1993). Different types of data collection and analysis can be used to cross-validate each other (Creswell, 2011; Stake, 2005). Furthermore, given the complex realities that are being studied, there is frequent overlap between data collection and data analysis (Stake, 2005; Yin, 2009). As described in the section Data Gathering, the data collection process for this study was iterative and therefore involved overlap with the data analysis; this was particularly true for the semi-structured interviews where there was initial analysis after each interview, which would then impact the subsequent interviews. Although this research has a single object of study, a company, there is a group level of analysis consistent with most of the organizational learning, learning organization, and knowledge management literature. This means there is an underlying assumption of group

53

homogeneity—both the responses to the Learning Organization Survey (Garvin et al., 2008) and the semi-structured interviews are analyzed as representative of the attitudes, opinions, and experiences of the company. Predictions of group homogeneity can be justified through organizational practices such as selection processes, socialization, and shared experience (Klein et al., 1994). The responses to the Learning Organization Survey (Garvin et al., 2008) were also compared across departments, in which case each department functioned as a sub-group within the larger company. For this inter-departmental comparison of results, the learning organization then is a multi-level construct because it assumes homogeneity within the teams but independence among the teams (Klein et al., 1994)—Edmondson’s Psychological Safety construct (1999) is similarly a multi-level construct because she compared results from interviews with different work teams within the same organization. Reliability and validity. Quantitative survey. According to Christensen et al. (2011), reliability “refers to consistency or stability of scores” (p. 143). The authors describe four type of validity: test-retest validity, equivalent-forms reliability, internal consistency reliability, and interrater reliability. The two types of reliability which are most relevant to the quantitative survey used in this study are test-retest reliability, which deals with the consistency of scores over time, and internal consistency reliability, which deals with the extent to which test items of a research instrument measure a single construct. In this study, test-retest reliability was established by administering the survey several times to different groups and individuals over a seven-week period, and by comparing the survey results from one week to the next in order assess the constancy of scores over time. Furthermore, in their published article about the survey, Garvin et al. (2008) present

54

benchmark scores representing survey results from other firms, which strongly suggests that both test-retest reliability and internal reliability have already been demonstrated. While many types of validity can be found in the literature on research methods, several scholars concur that construct validity is the most comprehensive and important notion of validity, subsuming many of the other types of validity (Christensen et al., 2011; Cronbach, 1989; Cronbach & Meehl, 1955; Grimm & Widaman, 2012; Messick, 1994). Christensen et al. (2011) state that “validity is based on evidence revealing that the target construct can correctly be inferred from the particular operations of measurement” (p. 146). Messick (1994), meanwhile, articulates a more explicit definition of construct validity as follows: “construct validity comprises the evidence and rationales supporting the trustworthiness of score interpretation in terms of explanatory concepts that account for both test performance and score relationships with other variables” (p. 7). Multiple sources of evidence of convergent and discriminant findings must be considered when assessing the construct validity of a study. According to Messick (1994), construct validity is the foundation for all other interpretations of test results, has both internal and external aspects, and is not only scientific but also rhetorical because it requires evidence and argument. Building on the work of several scholars—including Cronbach (1989), Cronbach and Meehl (1955), and Messick (1994)—Grimm and Widaman (2012) consider construct validity “as an overarching idea with two broad aspects—internal validity and external validity—each of which has more specific components” (p. 621). They state that: The internal validity of a test places the focus on the items that make up the test and how they are theoretically derived, relate to each other, and relate to the underlying construct(s) and whether they have the same meaning and difficulty for people with different background characteristics. (Grimm & Widaman, 2012, p. 624)

55

Internal validity deals therefore with the development and coherence of the test instrument itself, minimizing systemic error and bias, and is related to internal consistency reliability. Content validity, a component of internal validity, refers to the extent that a measure represents all facets of the given construct—it is frequently evaluated by a panel of expert judges; another component of internal validity is dimensionality, which refers to the relationship between test items and theoretical construct dimensions (Grimm & Widaman, 2012). The second broad aspect of construct validity, external validity, deals the extent that results can be generalized to other contexts. Like internal validity, it is considered to have several components including: criterion-related validity (which is composed of concurrent, predictive, and postdictive validity), convergent and discriminant validity, and change validity (Grimm & Widaman, 2012). Criterion-related validity deals with how the results of the test instrument correlate in expected ways with other previously validated test instruments which may have been implemented in the past, present, or future (Grimm & Widaman, 2012; Messick, 1989). Convergent validity is closely related to criterion-related validity because it deals with the extent to which test results are expected to correlate with other measures of the same construct; conversely, divergent validity deals with how results are expected not to correlate with measures of different constructs (Grimm & Widaman, 2012). Change validity deals with patterns of change over time - for example, expected growth curves (Grimm & Widaman, 2012). Building on the work of Messick (1994), Grimm and Widaman (2012) assert that “thorough examinations of the interpretations and consequences of test scores” are also part of external validity (p. 632). There are two main sources of potential invalidity which can undermine construct validity. First, construct underrepresentation occurs when the measurement is too narrowly

56

focused and therefore misses key dimensions or facets of the construct; second, construct irrelevant variance, the opposite of underrepresentation, occurs when the measurement is too broad, and hence includes variance which may be associated with other constructs, and/or the measurement includes method variance resulting from biases, or other factors irrelevant to the construct of interest (Messick, 1994). Determining what is relevant or irrelevant variance, however, can be a complicated question to address; according to Messick, it is a matter of how compelling the evidence and arguments are which attest that the source of variance is a “relevant part of the focal construct” as opposed to lending credibility to a rival hypothesis (p. 6). Like internal reliability, internal validity of the Learning Organization Survey (Garvin et al., 2008) can be established partly by the benchmark scores and quartiles provided by the authors which allow the researcher to compare results and assess the extent to which the scores lean to extremes versus the median scores for each item, and composite item. That the survey was developed by three experts in the field of the learning organization helps to establish content validity; and this is further reinforced by the review of the relevant literature in Chapter 2, which suggests that the survey captures important facets of the learning organization construct. Furthermore, as described in more detail in the Specific Design Elements section, the researcher went through several versions of the translated survey, had each version reviewed by native French speakers, and carefully examined the wording of each question with the e-commerce manager, and the human resources manager of the company, to ensure the participants would understand the survey items in a manner consistent with how native English speakers would understand the original version. Due to a paucity of well-accepted measurement tools in the learning organization literature, as noted in the literature review, certain aspects of external validity, such as criterion-

57

related, and convergent/divergent validity are difficult to establish for the Learning Organization Survey (Garvin et al., 2008). Because the objective of the survey is descriptive rather than causal, it is not necessary to establish predictive validity. Aspects of external validity, however, were established thanks to a thorough examination of the interpretations and consequences of the survey results carried out with the representatives of the company which allowed the researcher to verify the results were meaningful from their point of view. Qualitative semi-structured interviews. A review of the literature dealing with qualitative research methods suggests there is disagreement over whether or not terms reliability and validity are appropriate even if there is broad agreement on the overall objectives of the verification strategies. Shenton (2004), referring to the work of Guba (1981), states that while dealing with issues of reliability and validity, many qualitative investigators have “preferred to use different terminology to distance themselves from the positivist paradigm” (p. 63). The terms developed by Guba (as described in Shenton, 2004, p. 64) are credibility (instead of internal validity); transferability (instead of external validity/generalizability); dependability (instead of reliability); and confirmability (instead of objectivity). In order to establish the credibility (internal validity) of a qualitative research study, Shenton (2004) proposes a number of practices such as the adoption of well-established research methods, gaining an early familiarity with the culture of participating organizations, random sampling, triangulation, ensuring honesty in informants when contributing to data, iterative questioning, negative case analysis, frequent debriefing with research supervisors, peer scrutiny of the research project, reflective commentary in the research report, member checks, thick description of the phenomenon under scrutiny, and the examination of previous research findings.

58

With respect to establishing transferability (external validity), Shenton (2004) stresses the importance of context – the researcher should strive to include relevant information to help other researchers determine the extent to which the findings may be transferable to other contexts. Relevant information includes the number of organizations taking part in the study and their locations; limits or restrictions on the type of people who contributed data; the number of participants involved, the data collection methods employed; the number and length of the data collection sessions; and the time period over which the data was collected (Shenton, 2004). Moreover, Shenton (2004) stresses that variations in results, contradictions, and inconsistencies should not necessarily call into question the trustworthiness of qualitative research, but may reflect “multiple realities” of differing contexts (p. 71). The researcher should constantly ask what produces differences or similarities. Shenton (2004) states that “it should be questioned whether the notion of producing truly transferable results from a single study is a realistic aim or whether it disregards the importance of context which forms such a key factor in qualitative research” (p. 71). In order to establish dependability (reliability), Shenton proposes the use of overlapping methods such as individual and group interviews. He further recommends reporting all the processes in the study in detail to facilitate repeating the study. In his view, the study should be viewed as a “prototype model” (p. 71). The report should include the research design and its implementation, the operational detail of data gathering, and a reflective appraisal of the project (p. 72). Finally, to address issues of confirmability (objectivity), Shenton (2004) proposes that the researcher take steps to “help ensure as far as possible that the work’s findings are the result of the experiences and ideas of the informants, rather than the characteristics and preferences of the

59

researcher” (p. 72). He also proposes triangulation to “reduce the effect of investigator bias” (p. 72) and an ongoing “reflective commentary” in which the researcher puts forward and analyses his or her own potential biases. Lastly, Shenton (2004) proposes a data oriented approach leaving an “audit trail” (p. 72) which shows how the investigator developed his or her recommendations from the data over the course of the study. Morse, Barrett, Mayan, Olson, and Spiers (2002), on the other hand, argue that the alternative terminology developed by Guba (1981) and expanded by Shenton (2004), and its subsequent influence on qualitative research practices, has created confusion in the field. These authors find weaknesses with some of the provisions mentioned above. They question, for example, the usefulness of member checks, which is when the researcher discusses findings with the participants of the study; they assert this might be a threat to validity due to the participants’ inability to correctly judge the analysis. The authors further argue that certain practices, such as the audit trail constitute “ad-hoc evaluation” (p. 7). Overall, they assert that there has been a shift since the 1980s in ensuring rigor “from the investigator’s actions during the course of the research, to the reader or consumer of qualitative inquiry” (p. 1). This implies there is not enough rigor embedded into the research design and implementation. Morse et al. recommend using the same terms of reliability and validity widely used and accepted in quantitative research; these authors further maintain that “strategies for ensuring rigor must be built into the qualitative research process per se” (Morse et al., 2002, p. 9). According to Morse et al. (2002), verification strategies to insure validity and reliability in qualitative research include: ensuring methodological coherence, sampling sufficiency, developing a dynamic relationship between sampling, data collection and analysis, thinking theoretically, and theory development. Ensuring methodological coherence should develop out

60

of the iterative nature of the research because the researcher goes back and forth between design and implementation in order to constantly seek congruence “among question formulation, literature, recruitment, data collection strategies, and analysis” (Morse et al., Mp. 10). This requires responsiveness and flexibility on the part of the researcher—i.e. if the data don’t conform to expectations, the researcher should be open to new ideas inspired by the unexpected data and/or be ready to adjust aspects of the study in ways that are coherent with the data rather than adhering rigidly to a priori theoretical notions. Morse et al. (2002) further assert that theory, rather than being adopted as a framework to guide the analysis, may be the outcome of the research process and/or be used as a template for “comparison and further development of the theory” (p. 13). To summarize, whatever terminology is used, the work of Shenton (2004) and Morse et al. (2002) both propose useful verification strategies and best practices for qualitative research. For this study, numerous measures were applied to ensure the credibility (internal validity) of the semi-structured interview protocol, including the use of well-established research methods, gaining familiarity with the culture of the enterprise before starting the formal interview phase, iterative questioning (which helps ensure methodological coherence), and member checks. Likewise, measures were adopted to ensure transferability (external validity) such as providing relevant information about the company context, the study participants, the data collection methods employed, the number and length of the data collection sessions, and the time period. These measures are described in more detail in the Data Gathering section. Overlapping methods of individual and group interviews were used which help establish dependability (reliability). Measures were also taken to ensure the confirmability (objectivity) of the study such as triangulating the results of the interviews with the quantitative survey results, engaging

61

in reflective commentary, and leaving an audit trail. Finally, I developed a flexible and responsive approach to the data and to the data collection process by constantly checking the coherency between the methods used and the data, and also by using the theory and frameworks from the relevant literature for comparison and further theory development rather than rigidly imposing the theory on the data. Summary of verification strategies for both the quantitative survey and the semistructured interviews. Taking into consideration both the quantitative survey and the semistructured interview protocol, the study adheres to a number of recommendations and good practices put forward by Pinsonneault and Kraemer (1993) for survey research, such as the use of a mixed methods approach to develop a more complete and detailed understanding of the results; i.e. the survey results help to complete the results of the semi-structured interviews and viceversa. Second, a robust percentage of the company’s representatives participated in the study (36%), and all levels of the organization were represented, along with a wide variety of roles within the company, which helps to establish a strong link to the unit of analysis (the company). Third, thanks to the face-to-face group-administered questionnaire method, a 100% response rate was achieved to the survey instrument. On the other hand, the study falls short on other recommendations put forward by Pinsonneault and Kraemer (1993), such as having a larger sample size (more than one organization) and applying systematic random sampling techniques for the survey respondents instead of a convenience sample. Specific Design Elements Learning organization survey. Garvin et al. (2008) cite three factors which have prevented the ideal of the learning organization from coming to fruition. First, they argue that prior discussions of the concept have been too idealistic and have lacked concrete advice for

62

practitioners; second, the literature has focused too heavily on the highest levels of the organizations, such as the CEO and senior management, without giving adequate consideration to lower levels where important organizational work also takes place; third, they maintain there has been a lack of standards and tools to assess that organizational learning was taking place (Garvin et al., 2008). These concerns echo Garvin’s earlier call in the learning organization literature for three things: meaning (a good clear definition), management (clear guidelines for practice), and measurement (Garvin, 1993/2000). The Learning Organization Survey (Garvin et al., 2008) and its accompanying article represent an attempt to address these deficiencies. As a group-level measurement tool, it is intended to measure learning which occurs in “an organizational unit of any size that has meaningful shared or overlapping work activities” (Garvin et al., 2008, p. 110); such a unit may be a department, a project, a service, or a division. Thanks to having administered the survey at various firms, the survey authors have compiled benchmark scores which not only enable an organization as a whole to compare its results to those benchmark scores, but also enable each organizational subunit to do so; and, moreover, these benchmark scores allow comparisons across subunits within the same organization. The authors assert that “the power is in the comparisons, not in the absolute scores” (Garvin et al., 2008, p. 110). The survey gathers information at a single point of time from a sample representing the total population; its goal is to get a precise measurement of employee perceptions in the form of opinions about their workplace. The items in the survey pertain to the multidimensional theoretical construct of the learning organization described in the relevant section in Chapter 2. The survey tool can therefore be considered a single cross-section descriptive survey (Christensen et al., 2011). For the purposes of this study, the survey was translated into French

63

(see Data Gathering section below). Two questions were added at the beginning of the survey to determine first the participant’s level in the company hierarchy (member of executive committee, manager, or employee), and, second, to determine in which department the participant worked (operations, sales and marketplace, human resources, e-commerce, or finance). The original version of the survey has 55 questions and the version used for this study has 57 questions because of the two added questions. Except for the two added questions, which capture attributes about the participant, and allow for comparison of results across subunits, the survey consists of Likert Items in the form of statements the participant is asked to evaluate, such as: if you make a mistake in this unit, it is often held against you. These items are paired with Likert Scales which are defined as a range of choices with a balance of positive and negative responses and a neutral or zero value (Christensen et al., 2011). There is some debate in the literature dealing with surveys about whether or not having a neutral or zero value is an advantage. While some consider having no neutral midpoint is an advantage because it forces participants to choose, Christensen et al. (2011) argue that because midpoints tend to draw an equal number from both sides, the overall distribution is not changed. Furthermore, forcing participants to make a choice might generate anxiety. For 47 of the 55 Likert Items, participants choose from among seven possible responses: highly inaccurate, moderately inaccurate, slightly inaccurate, neither accurate nor inaccurate, slightly accurate, moderately accurate, and highly accurate. For the last eight Likert items, participants choose from among five possible responses: never, infrequently, sometimes, often, always. In order to reduce the potential for response set bias, which is when respondents mark the same area of rating scale regardless of the question (Christensen et al., 2011, p. 352), reverse wording is sometimes used. For example, the statement “in this unit, it is easy to speak up about what is on

64

your mind” is followed by the statement “if you make a mistake in this unit, it is often held against you”. If a participant chooses “highly accurate” as response for the first statement, he or she would probably choose “highly inaccurate” for the following statement because the two statements describe very different, even opposing company cultures. In the scoring of the questionnaire responses, some items are therefore also reversed (i.e. the response “highly accurate” is a high score value for the first statement whereas it is a low score value for the second). The survey is designed to answer two questions about the organization or organizational unit: to what extent it functions as a learning organization; and what the relationships are among the factors which affect the organizational learning (Garvin et al., 2008). The survey is built on the theory that there are three major factors which must be present to some extent for organizational learning to occur: a supportive learning environment; concrete learning practices; and leadership that reinforces learning. Each of the three factors is referred to as a “building block” and is broken down into subcomponents under which the questions are grouped; these are summarized below. To complete the survey, participants rate how accurately the statements match their perceptions of the organization (or subunit) they work for, and its learning culture. Thanks to the multichotomous rating scale, the survey measures both the direction (positive or negative), and the strength of the respondent’s opinion about the statement (slightly, moderately, highly, etc.). The survey is designed so that several members from the same unit can take the survey, and so that their scores can be averaged in order for the collective scores to be analyzed on a group level. A complete example of the original survey can be found in Appendix A1; a complete example of the survey translated into French can be found in Appendix A2.

65

Building Block 1, Supportive Learning Environment (Garvin et al., 2008), includes the following subcomponents: 

Psychological Safety: This subcomponent is largely based on the work of Edmondson (1999, 2002) who maintaints that in order to learn effectively, employees need to feel comfortable expressing their views without fear of retribution. Edmondson (1999) suggests that psychological safety can impact team learning and performance.



Appreciation of Differences: The main idea of this subcomponent is that different points of view can stimulate new ways of thinking and prevent organizations from getting stuck in routines and institutional inertia. It is therefore desirable that an atmosphere of tolerance and curiosity towards differences exist in the enterprise.



Openness to New Ideas: This subcomponent is about the extent to which employees are encouraged to take risks. Double loop learning (described in Chapter 2, Organizational Learning section) means employees go beyond just trying to improve effectiveness in order to question underlying assumptions. Edmondson (2002) uses the terms “radical” versus “incremental” learning to describe the same phenomena. Awareness of opposing ideas and a willingness to consider them is essential for meaningful organizational learning.



Time for Reflection This subcomponent is closely linked to the Openness to New Ideas subcomponent because double loop (or radical) learning requires managers and/or employees to step

66

back from their daily routines to reflect on the underlying assumptions, and to engage in real discussion. To be constantly under stressful productivity time constraints means employees cannot question the way things get done—they do not have adequate distance from their work to consider long term solutions to problems, or to conceive of entirely new ways of doing things. Building Block 2, Concrete Learning Processes and Practices (Garvin et al., 2008), includes the following subcomponents: 

Experimentation: This subcomponent is about the extent to which rigorous and systematic processes are in place for developing and testing new ideas, products and services. For example, the enterprise may frequently try out new products using prototypes and can fall back on established processes for assessing how they perform.



Information Collection: This is about the extent to which the organization has clear processes for gathering intelligence on consumers, market trends, technological trends, and competitors. This also implies that the enterprise has methods and procedures for comparing its performance to competitors or other well-respected companies in the sector.



Analysis: This subcomponent is about the organization’s disciplined analysis, and interpretation of the information gathered. It relates to the extent to which discussions about the interpretation of the information gathered allow for differences of views and questioning underlying assumptions.

67



Education and Training: This deals with how the company develops employees, both new and old, and the extent to which training is valued. Whether or not periodic training is offered to all employees or only to new employees is a key question. Some companies have little or no formal training programs.



Information and Transfer: This subcomponent deals with how the organization shares information—whether between individuals, among groups, or across the entire organization—and how it deals with the flow of information. For example, some departments might organize information meetings with experts from other departments, or from outside the organization. Or the organization might regularly share information with outside experts, customers, and/or supplies. Building Block 3, Leadership that Reinforces Learning (Garvin et al., 2008), has no

subcomponents and is the shortest section of the survey covering only eight questions. It tries to capture the behavior of the organization’s leaders as perceived by the participants completing the survey. The building block deals with the extent to which leaders are perceived to engage in behaviors that will encourage learning such as asking good questions, actively listening, prompting dialogue and debate, demonstrating openness to points of view different than their own; and by showing willingness to invest resources and time in problem identification, problem solving, and reflection about making improvements on past performance. Semi-structured interview protocol. I developed an interview protocol for this study by combining elements of the causal mapping methodology proposed by Ambrosini and Bowman (2001), with the knowledge-in-practice framework developed by McIver et al. (2013).

68

Semi-structured, open-ended interviews were conducted with both individuals and small groups immediately after each participant completed the survey described above. The first part of this section describes the causal mapping methodology and the second part deals with the knowledge-in-practice framework. Causal mapping methodology. Ambrosini and Bowman (2001) propose a causal mapping methodology as a tool to operationalize tacit knowledge. Citing Barney (1991), they assert that tacit knowledge is considered crucial to the development of a firm’s sustainable competitive advantage; however, they argue that the construct has “resisted operationalization” (p. 811). Consequently, according to them, many of the assumptions about it have not been adequately supported by empirical research—for example, the assumption that tacit knowledge is difficult, if not impossible, to imitate, transfer or substitute (Ambrosini & Bowman, 2001). They moreover conclude that the term “tacit skills” is preferable to “tacit knowledge”, particularly “within the context of the resource-based view of the firm” because the construct deals with know-how and practice rather than abstract knowledge (Ambrosini & Bowman, 2001, p. 814). Ambrosini and Bowman (2001) suggest that while tacit skills are embedded in the individual, attached to the knower, and deeply rooted in contexts (i.e. specific professions, technologies, markets, activities or work groups), tacit skills are not necessarily unknowable. Furthermore, along with Nonaka (1994, 1997/2001), Spender (1993, 2014), and Cook and Brown (1999), Ambrosini and Bowman (2001) see tacit knowledge and explicit (or scientific) knowledge as being part of a larger whole rather than as separate and opposing constructs. Citing the work of Polyani (1966), Ambrosini and Bowman (2001) contend that what is explicitly knowable is only a part of what people know—the part that can be expressed in words.

69

According to this view, tacit knowledge is considered more fundamental than explicit knowledge—it is a base of raw knowledge from which explicit knowledge stems. One of the most important theoretical notions in the causal mapping methodology is that tacit knowledge (defined as tacit skills) is a nuanced concept. The authors propose four degrees of tacitness as follows: High A: Deeply ingrained tacit skills B: Tacit skills that can be imperfectly articulated C: Tacit skills that could be articulated D: Explicit skills Low

Figure 1. Degrees of tacitness (Ambrosini & Bowman, 2001, p. 816) According to Ambrosini and Bowman (2001), A and D in Figure 1 can be considered anchor points. Deeply ingrained tacit skills (A) are inaccessible to knowers because they cannot be communicated; while explicit skills (D) can easily be communicated, codified and shared. The causal mapping methodology focuses primarily on the points in between (B and C). Tacit skills that can be imperfectly articulated (B) may not be expressed effectively “through normal use of words” (Ambrosini & Bowman, 2001, p. 816) but they can be expressed through storytelling and metaphors. Tacit skills that can be articulated (C) may be elicited by asking the right questions. The authors suggest these skills (C) may have, in fact, been acquired explicitly but have become tacit over time; this implies they are tacit mainly because people don’t think consciously about what they do, perhaps simply because no one has asked them to (Ambrosini & Bowman, 2001). Cook and Brown (1999) in their work on the epistemologies of knowledge and knowing use the metaphor of riding a bicycle to explain different types of knowing. The metaphor also works well to illustrate the different degrees of tacitness in Ambrosini and Bowman’s (2001) 70

methodology. Knowing how to ride a bike doesn’t necessarily mean you can explain to someone else how to do it. For someone who can ride a bike but can’t effectively talk about it or explain how to do it, riding a bike would correspond to A in Figure 1. For someone who can speak “imperfectly” about riding a bike, for example, by saying that gliding down a hill is like flying, or by telling a story about a recent near accident due to loss of balance, riding a bike would correspond to B in Figure 1. For someone who could answer questions about riding a bike, such as “When do you shift gears?”, or “How do you slow down?”, riding a bike would correspond to C in Figure 1. Finally, for someone who could actually give explicit instructions and/or write a manual about how to do it, riding a bike would correspond to D in Figure 1. In the work of Cook and Brown (1999), the bicycle riding metaphor illustrates knowing through practice which feeds into other types of knowledge and is part of a dynamic, generative process. Their epistemology of knowledge and knowing is very consistent with Ambrosini and Bowman’s philosophical stance on knowledge—that subjective reality based on experience is not separate from objective reality. Referring to Berger and Luckman (1966), and citing Crotty (1998), Ambrosini and Bowman state that “all knowledge and therefore all meaningful reality as such, is contingent upon human practices, being constructed in and out of interaction between human beings and their world, and developed and transmitted within an essentially social context” (2001, p. 817). The causal mapping methodology makes use of semi-structured interviews designed to ask the “right questions” in order to elicit tacit skills (such as those corresponding to C in Figure 1), and to get participants to tell stories and/or make use of metaphors in order to talk indirectly (or “imperfectly”) about tacit skills (such as those corresponding to B in Figure 1). Telling stories may reveal tacit skills because people tend to “frame their experience in stories”

71

(Wilkens & Thompson, 1991, as cited in Ambrosini & Bowman, 2001, p. 820); and because stories tend to be less filtered than responses to traditional interview questions; the authors recommend asking participants to tell one negative and one positive story when talking about things which have happened recently and which may be related to the construct of interest. Likewise, metaphors are useful in revealing tacit skills when explicit language is not available because of complex or ambiguous experience. Nonaka (1991/2007, 1994) stresses the importance of using metaphors to capture tacit skills in order to more effectively spread knowledge throughout the enterprise. In sum, the causal mapping method “is an indirect way of surfacing tacit skills” (Ambrosini & Bowman, 2001, p. 825). Ambrosini and Bowman assert that the method “will be fragmented, not comprehensive, partial, and biased but it should provide some insights to both participants and researchers into tacit skills and organizational success” (2001, p. 825). In practice, the method consists of eliciting factors related to organizational success— these become the base constructs which the mapping process is designed to uncover in the quest to reveal tacit skills. Once the researcher has established based constructs related to organizational success, he or she then begins the mapping process to discover the causes or enabling factors of these constructs. This mapping process consists of asking questions and eliciting stories, as illustrated in Figure 2.

72

Base construct related to organizational success

Factor A

Factor C

Factor B

Causes of Factor A

Example of cause or enabling factor

Interviewer Questions:

What causes A to happen?

How does A happen? Can you give an example?

Story about how that happened recently

Can you tell a story about the occurrence of A?

Figure 2. Researching tacit skills: a summary of the proposed method (adapted from Ambrosini & Bowman, 2001, p. 823) Having identified factors related to the construct of interest (such as organizational success) either in pre-interviews, or early in the interview process, the interviewer starts the mapping process by asking questions about the factor such as: “What causes A to happen?”. When a response is elicited, the interviewer tries to go deeper by asking questions such as: “How does A happen?”, and, “Can you give me an example?” When this example is elicited, the interviewer attempts to get the participants to tell stories about an occurrence of A. The two initial interviews with the e-commerce manager, and the meeting with the company representatives, which preceded the start of the formal in-company individual and group interviews, allowed me to better know the company and its top managers, and to increase

73

their confidence in the process. Moreover, I was able to elicit base constructs related to organizational success such as commercial success, innovation, high-performance, and employee and client development. Knowledge-in-practice framework. The knowledge-in-practice framework is a contribution to the field of knowledge management developed by McIver et al. (2013); these authors define knowledge management as “a set of management activities aimed at designing and influencing knowledge creation and integration (including processes of sharing knowledge)” (p. 597); their framework “describes knowledge characteristics of work practices along two dimensions: tacitness and learnability” (p. 597). McIver et al. challenge what they perceive to be an assumption underlying much of the knowledge management theory and literature—that knowledge management practices necessarily lead to improved performance. On the contrary, they suggest that mismatches sometimes occur between knowledge management practices and the contexts they are supposed to apply to, and that mismatches can result in a regression in performance. As a tool to describe the “underlying knowledge structure” (degrees of tacitness and learnability) of tasks in organizational work settings, the framework developed by McIver et al. is meant to help researchers and practitioners design more effective knowledge management planning and application strategies. Citing Brown and Duguid (2001), McIver et al. (2013) define the term practice to mean “the way in which work gets done and knowing how to do it” (p. 599); the term applies in a broad sense to work associated with jobs and professions. McIver et al. consider the tacitness of a practice as unobservable aspects of it, and they concur with Ambrosini and Bowman (2001) that there are degrees of tacitness, from low to high, which partly define a practice. They also suggest that all tacit knowledge is not necessarily difficult to learn, while simultaneously arguing

74

that not all information (facts and data) is “comparatively easy to learn” (McIver et al., 2013, p. 600). The boundary between explicit and tacit is knowledge is therefore blurred similar to what we find in the theory underlying causal mapping. The authors’ dimension of learnability meanwhile includes both notions of tacitness and information: “The learnability of a practice consists of the type and amount of effort, study, accumulated comprehension, and expertise required to understand the information and know-how involved in work activities” (McIver et al., 2012, as cited in McIver et al., 2013, p. 600). The extent to which jobs might be easy to learn or hard to learn depends on a number of characteristics including how complex the knowledge is, whether or not there is causal ambiguity, how much inconsistency there is, and/or how much accumulated information and experience is required to perform the job (McIver et al., 2013). Based on the two dimensions described above, the authors propose four practice types, as presented in Figure 3. 1. Enacted information = high learnability, low tacitness (example: a grocery store cashier) 2. Accumulated information = low learnability, low tacitness (example: a tax auditor) 3. Apprenticed know-how = high learnability, high tacitness (example: a roofer or carpenter) 4. Talent and intuitive know-how = low learnability, high tacitness (example: an artist or an athlete) Figure 3. Knowledge-in-practice for organizational work (adapted from McIver et al., 2013, p. 601). To illustrate the first practice type, enacted information, the authors propose the example of a grocery store cashier. They consider this type of practice easy to learn because it only involves a small amount of tacit knowledge (experience-based expertise and know how); explicit and codified knowledge is the main ingredient. Furthermore, the practice tends to remain stable over time. This characterization provides a good explanation of why this type of work is being

75

increasingly automated. Other examples given by the authors of this practice type include jobs such as a bank teller, a mail carrier, or a medical records clerk. The second practice type, accumulated information, is illustrated by the example of a tax auditor. Similar to the first type, there is little tacit knowledge involved—the main difference lies in the amount and complexity of information. The authors state that “the knowledge involved in these types of practices can be observed, codified, and disaggregated from its context, but it is more difficult to learn because expertise requires extensive mastery of large amounts of information” (McIver et al., 2013, p. 601). The studies and training required to become a tax auditor mainly consist of mastering and assimilating the vast amounts of codified information and in knowing to what contexts it applies. Other examples of this practice type given by the authors include professions such as engineering, financial analysis, and editing. To illustrate the third practice type, apprenticed know-how, the authors give the example of roofing as well as other skilled labor jobs including carpentry, plumbing, and radiology. They characterize the practice as easy to learn but high in tacitness. This implies that learning the activity requires a great deal of practice and experience. Even though the know-how is largely unobservable and difficult to codify, “learnability is high … because the connections between required actions and performance are consistent and comparatively simple to replicate and the steps needed to learn how to perform these actions can be readily identified” (McIver et al., 2013, p. 601). The authors characterize the fourth practice type, talent and intuitive know-how, as “artistry”. This type of practice is considered to be high in tacitness and difficult to learn, and rarely transferable. This is because the required know-how is complex and constantly evolving. The authors state it can only be developed “through idiosyncratic experience” (McIver et al.,

76

2013, p. 602). This type of practice is associated with unique talent—hence the following cases are given as examples to illustrate it: “famous artists, world-class athletes, gifted creative teams, or scientific geniuses” (McIver et al., 2013, p. 602). Integrating the knowledge-in-practice framework and the causal mapping methodology. As previously noted, there are numerous areas where the work of Ambrosini and Bowman (2001) and the work of McIver et al. (2013) overlap and/or complement each other. For example, there is a strong emphasis on work practices in the theory underlying both the causal mapping methodology and the knowledge-in-practice framework. Likewise, a shared understanding can be found of tacitness as a nuanced construct which can be expressed and analyzed in varying degrees. Furthermore, there is a shared philosophical stance that tacit and explicit knowledge are not two worlds apart but should be understood as dynamic aspects of knowledge in a larger sense. The knowledge-in-practice framework can be easily adapted to the causal mapping methodology because high performance (a key construct in the former) can be considered an organizational success factor (a key construct in the latter). The summary of the knowledge-inpractice framework found in Appendix B in table format presents each knowledge-in-practice type in the first column, and then breaks down horizontally each type for analysis according to the following questions in Figure 4. 

What is high performance?



How is high performance achieved?



What needs to be known?



How does knowing take place?



How is knowing applied?



Examples of types of work?

Figure 4. Knowledge-in-practice framework questions (adapted from McIver et al., 2013, pp. 605-606).

77

For this study, the above questions were integrated into the causal mapping methodology in the following manner:

Base construct related to organizational success

Factor A (High performance)

Factor C

Factor B

Cause of high performance

Interviewer Questions:

What is high performance?

Knowledge required for high performance Story about recent example of high performance for a specifc job or taks

How is high performance achieved?

What needs to be known?

How does knowing take place/how is knowing applied?

Figure 5. Integrating the knowledge-in-practice framework into the causal mapping methodology (adapted from McIver et al., 2013; and Ambrosini & Bowman, 2001). Thus, the causal mapping methodology is enriched by the knowledge-in-practice framework. The questions from the table presenting the framework are used to uncover tacit knowledge while simultaneously revealing the knowledge structure of company practices (i.e. jobs and tasks). This allows for a potentially deeper and more complex analysis of the data. The exact wording from the knowledge-in-practice questions, however, needs to be adapted to the interview protocol because questions such as “How does knowing take place?”, or “How is knowing applied?” are not necessarily clearly understood by participants. A question such as

78

“Can you give me an example of how someone applies his knowledge effectively on the job?” is a more effective way of eliciting the same information. Data Gathering Engaging with the company. I initially approached the e-commerce manager of the study company a little less than one year before starting the formal interview process, to discuss conducting a study related to employee turnover and organizational learning. We had met previously at an informal occasion where our conversations about the company he worked for led me to think it would be a good match for the type of research I wanted to conduct. The first meeting we had allowed me to gain an overview of the company, the sector it operated in, its history, and its current situation. Five months later, we had a second meeting to discuss the match between the company objectives and my research objectives. Based on our discussions, he organized a meeting one month following, which, this time, included himself, the newly hired chief financial officer, the newly hired human resources director, and me. They were all receptive to the idea but expressed more interest in organizational learning and knowledge sharing than employee turnover. Several days later, I drafted a modified outline of the study which then became the rough template for moving forward. This template outlined a study which involved a survey to be given to all the employees, and semi-structured interviews to be carried out with about half the staff—individual interviews for members of the executive committee and top managers, and small group interviews for the other employees. The goal of the survey would be to assess the extent to which employees perceived the company’s practices as consistent with the concept of the learning organization, and to assess perceived strengths and weaknesses (relative to the learning organization concept). The goal of the semi-structured interviews would be to uncover and to assess tacit knowledge with a view to gaining a better

79

understanding of this knowledge and how to manage it. It was agreed that the survey would be implemented as a first step and would be sent out by the e-commerce manager to all the employees of the firm via e-mail with a link to complete it online. After this, the human resources manager was to organize the individual interviews followed by the group interviews. The survey which I chose for the study, the Learning Organization Survey (Garvin et al., 2008), is described in the relevant section above. It is available to anyone online for free at the following link: https://hbs.qualtrics.com/jfe/form/SV_b7rYZGRxuMEyHRz. Once anyone has taken the survey, the platform immediately analyzes the results and provides an individual score which can be compared to benchmark scores published in the accompanying Harvard Business Review article. However, I could not use the survey in that format for two reasons. Firstly, the survey had to be translated into French for the employees of the company to understand it, and, secondly, I had to be able to aggregate results on the company, and on department levels to allow for comparing across departments. I therefore sought and obtained permission to do this from the authors. Since I was not able to get any information on the scoring, however, I had to work from the benchmark scores and deduct how much each possible response was worth in order to create my own scoring sheet. I then proceeded to translate the survey myself and had it verified by two native French speakers. When a beta version of the survey was shared, however, with the e-commerce manager, and the human resources manager at the company, some difficulties arose. The human resources manager was concerned that some of the questions were not readily understandable to the employees despite the translation. Furthermore, she had concerns that employees would not cooperate reliably to complete an online survey with 57 questions. The company had apparently experienced difficulty in the past trying to get employees to complete online surveys that were

80

much shorter than the Learning Organization Survey (Garvin et al., 2008). She considered that the costs in time and mental effort for the employees would discourage them from completing the survey, and that consequently the completion rate would be low. The e-commerce manager, who was more optimistic that the difficulties could be surmounted, suggested the company organize a thirty-minute session during which all the employees of the company would take a break from their daily work and complete the survey together. Later, however, due to the fast pace of work, productivity demands, this suggestion was judged to be impracticable. In order to get around what was becoming an impasse, I proposed the group-administered questionnaire method (Christensen et al., 2011). This involved modifying the data collection strategy so that a paper version would be given to each interview participant just before the semistructured interviews; I would hand out the surveys myself and have the participants complete it during the first 20 minutes of the scheduled interview session. While this meant that the survey would not be sent to the entire population as originally planned, the sample size of 42, which represented 36% of the company employees, would be robust enough to reliably represent the population. Furthermore, this method had some advantages when it came to the quality and percentage of survey responses. First, the face-to-face contact became an opportunity for me to explain the objectives of the study, and to insist on the confidentiality of the procedures and results. This helped build trust among the participants which, according Christensen et al. (2011), contributes to the quality of responses. Second, I was nearly guaranteed to get a 100% response rate because I would remain with the participants while they completed the surveys, and then immediately collect them. The company agreed to this strategy, and the human resources manager planned the survey-interview sessions in meeting rooms so that the

81

participating employees and managers would be temporarily away from their daily tasks but still at work. Because of scheduling challenges, the initial plan to conduct the individual interviews with members of the executive committee and some top managers before doing the group interviews was modified in favor of chronologically overlapping individual and group interviews. Five group interviews and five individual interviews were scheduled to take place in the company over a period of three weeks. One individual interview was then added during the fifth week, and two individual interviews were then added during the seventh week; the time interval separating these last two interviews from the first cluster of interviews was due to endof-year holidays and vacations. Carrying out the survey-interview sessions. Thirty-four employees participated in the group interviews—the size of the groups ranged from five to nine. Employees represented different job types and departments. For example, the first group included seven employees: two web developers, one employee from human resources, a digital growth manager, a business developer, a web traffic analyst, and someone from the fraud section of the finance department. For complete interview transcripts which include lists of individual job titles for each participant, see Appendices C1 and C2. Seven people participated in the individual interviews including four members of the executive committee (a co-founder and current head of photography, the head of commercial development and strategy, the chief financial officer, and the e-commerce manager), and three top managers (the human resources manager, a brand manager, and the head of client relations). As stated above, the scheduling of the two types of interviews overlapped— an individual interview was followed by a group interview, followed by two individual interviews, followed by a group interview, et cetera. Interviews were spread out so that not more

82

than two sessions (individual and group) were scheduled on the same day. Interviews lasted from one hour to one hour and thirty minutes including 20 minutes to complete the survey. Table 1. Sequence of initial interviews and formal survey/interview sessions Step

Who was involved

Timing

What was accomplished

Initial interview one

E-commerce manager

Eleven months prior to formal interviews

Gained an overview of company history, current situation, and possible areas of research.

Initial interview two

E-commerce manager

Six months prior to formal interviews

Discussed match between research objectives and company objectives/

Initial meeting with company representatives

E-commerce manager, human resources manager, and chief financial officer

Five months prior to formal interviews

Revisited match between research objectives and company objectives. Discussed implementation of study, company structure, and elicited key organizational success constructs.

Start of formal in-company individual and group interviews Individual survey + individual interview one

Co-founder and director of photography

Week one

Reviewed company history, management philosophy, and elicited key organizational constructs and success factors.

Group survey + group interview one

Seven employees from different levels and departments

Week one

Causal mapping based on key organizational constructs and success factors, application of knowledge-in-practice framework.

Individual survey + individual interview two

Human resources manager

Week two

Checked and reinforced management philosophy, and key organizational organizational constructs and success factors.

Group survey + group interview two

Five employees from different levels and departments

Week two

Causal mapping based on key organizational constructs and success factors, application of knowledge-in-practice framework.

Individual survey + individual interview three Group survey + group interview three

Head of commercial development and strategy

Week two

Checked and reinforced key organizational constructs and success factors.

Six employees from different levels and departments

Week two

Causal mapping based on key organizational constructs and success factors, application of knowledge-in-practice framework.

Individual survey + individual interview four Group survey + group interview four

Chief financial officer

Week three

Checked and reinforced key organizational constructs and success factors.

Nine employees from different levels Week three and departments

Causal mapping based on key organizational constructs and success factors, application of knowledge-in-practice framework.

Group survey + group interview five

Nine employees from different levels Week three and departments

Causal mapping based on key organizational constructs and success factors, application of knowledge-in-practice framework.

83

Individual survey + individual interview five

E-commerce manager

Week five

Individual survey + individual interviews six and seven

Brand manager, and after sales service manager

Week seven

Follow up interview one

E-commerce manager, and human resources manager

Follow up interview two

E-commerce manager, and human resources manager

Five months after the last formal survey/interview sessions Six months after the last formal survey/interview sessions

Checked and reinforced key organizational constructs and success factors, checked and reinforced causal mapping, checked and reinforced application of knowledge-inpractice framework Checked and reinforced key organizational success factors, checked and reinforced causal mapping, checked and reinforced application of knowledge-in-practice framework Discussed findings and recommendations.

Discussed potential applications of findings and recommendations for future training program.

Data Analysis Survey score analysis. According to Pinsonneault & Kraemer (1993), for descriptive or exploratory survey research, analysis of test scores generally consists of developing the marginal and cross-tabulations for the variables and applying descriptive statistics such as means and medians. As previously explained in the subsection Engaging with the company, the scoring system for the questionnaire was deducted by taking the benchmark scores provided by the authors, checking the perfect score for each subcomponent, and then dividing by the number of items in the subcomponent. By completing the survey myself several times and checking my own scores, I was able to verify my deducted calculations. For example, a perfect score is 100 for Psychological Safety, which is the first subcomponent of Building Block 1, Supportive Learning Environment. Because this subcomponent includes five statements, the highest value for an item is twenty. Since there are seven possible responses to the statement, twenty is divided by seven to determine the lowest value for an item which is 2.86. The second lowest value is then determined by doubling this figure, the third lowest value by tripling this figure, and so on. The spread of values therefore for an item with seven possible responses is: 2.86,

84

5.72, 8.58, 11.44, 14.3, 17.16, and 20. If the statement has a positive value (such as “in this unit, it is easy to speak up about what is on your mind”), the response “highly accurate” (très juste) is attributed a value of twenty, and the response “highly inaccurate” (pas juste du tout) is attributed a value of 2.86, while the midpoint response, “neither accurate nor inaccurate” (ni vrai ni faux) is attributed a value of 11.44. Hence, someone choosing midpoint responses for each of the five items of the subsection would get a score of 11.44 x 5 = 57.2. This score would put the individual in the bottom quartile (31-66) of the benchmark scores compiled by the authors. A simple set of descriptive statistics including the group mean, median, and standard deviation was then generated for all the participants and broken down into collective results for samples of subunits of the enterprise (i.e. different departments). A complete scoring data sheet developed by me can be found in Appendix D, and a complete breakdown of the benchmark scores compiled by the survey’s authors can be found in Appendix E. The median scores from the test results can be compared to the median benchmark scores provided by the authors. Analysis of semi-structured individual and group interviews. As presented in Table 1, the data collection process for the study was iterative and hence allowed for overlap between collection and analysis. According to Morse et al. (2002), “collecting and analyzing data concurrently forms a mutual interaction between what is known and what one needs to know” (p. 12). The two initial interviews with the e-commerce manager, and the initial meeting with the company representatives were crucial for establishing research objectives, and a study design that would be workable, and mutually beneficial for both me and the company. Furthermore, the initial meetings allowed me to gain familiarity with the organization, its history, and its management culture before starting the causal mapping and knowledge-in-practice framework interviews. Finally, the initial interviews and meeting were useful because they

85

helped me to develop some initial ideas of important constructs and factors for organizational success which could then be explored during the formal interviews. The individual causal mapping and knowledge-in-practice interviews were used to focus on the deducted constructs and organizational success factors, to elicit other possible factors, to explore possible causes of those factors, and to solicit specific examples. The interviews were also used to examine dimensions of work practices such as notions of high performance, the complexity of the knowledge required, and how learning took place. Individual interviews furthermore yielded information and points of view about aspects of the company specific to the positions and personalities of the individuals interviewed. The first individual interview, with the co-founder and director of photography, focused on the history of the founding of the company, its management philosophy, how this philosophy has developed overtime, how it has been implemented, the company’s decision making processes, and ways in which the company is innovative. This individual interview also briefly addressed the key competitive advantages of the enterprise, and its potential threats. The second individual interview, with the human resources manager, was an opportunity to get a different perspective on some of the same issues addressed in the first interview; the human resources manager is relatively new to the company, is not a member of the executive committee, and, unlike the co-founder, is deeply involved in day-to-day management and staffing concerns. Her perspective was therefore sought on the management philosophy and its implementation, and the company decision-making processes. Notions of high performance were likewise addressed with the human resources manager. The third individual interview was with the head of commercial development and strategy, who has been with the company for about a year and a half. This was an opportunity to

86

revisit, from another point of view, the questions of innovation, competitive advantage and threats; he is in charge of implementing the new business model, and, moreover, he previously worked for a client of the company. Notions of commercial success were explored during this interview. The forth individual interview was with the chief financial officer, who has also been with the company for a little over one year; during this interview, the concept of employee growth and development was briefly addressed. The fifth individual interview, with the e-commerce manager, happened after most of the other individual and group interviews had already transpired, and was used to clarify some aspects of the company’s activities and structure. For example, clarity was sought pertaining to the relationships between activities such as marketplace, marketing, e-Store, and traffic. Also questions were addressed about how certain positions worked together, such as brand managers, and project managers. Finally, this interview was used to discuss and verify dimensions of the knowledge-in-practice framework that had come up during the group interviews such as notions of high performance, complexity of knowledge required, and learning. The last two individual interviews, with a brand manager and the head of client relations, respectively, were originally scheduled together in one group interview. However, based on the data that had been gathered and analyzed thus far from the previous group interviews, I judged it more useful to interview each of them separately. Following the suggestion of my research supervisor, these two final individual interviews focused particularly on daily tasks and problem solving. Each group interview, which included anywhere from five to nine employees from various levels and departments, followed an individual interviews until group interview five, when this pattern was reversed because of staff availability constraints. The group interviews were used to explore key organizational constructs and success factors identified in the initial

87

interviews and during the formal individual interviews; key organizational constructs included: innovation, commercial success, good relations with the client, and employee and client development. As per the causal mapping method, employees were asked to define these organizational constructs, to talk about factors of these constructs, their causes, and, whenever possible, to provide examples, narratives and metaphors with a view to uncovering tacit knowledge (skills). After consultation with my dissertation supervisor, the later interviews, particularly group interview five, included more direct questions about the participants’ daily activities. This is because many of the issues and themes arising in the group interviews were recurrent which suggested the data set was near completion (i.e. saturation); however, data on tacit skills remained sparse. In an attempt to elicit more tacit skills, questions were asked about what a typical day consisted of, and about recent problems and ways of facing them. The group interviews were used to explore dimensions of the knowledge-in-practice framework by asking employees how they defined high performance in their jobs, what type of knowledge was required, and how learning transpired. Two follow up meetings were organized with the e-commerce manager and the human resources manager after the formal survey/interviews were complete, and the data analyzed. One meeting took place five months after the interviews, and the second meeting took place six months after the interviews. Initial findings and recommendations were discussed along with a future company training initiative related to the results of the study. The transcribed interview texts were analyzed for common themes and for tacit skills elicited or revealed indirectly. They were also analyzed using the knowledge-in-practice framework; according to the descriptions of practices and relevant examples provided by the participants, jobs and roles were classified according to the knowledge-in-practice types

88

developed by McIver et al. (2013): enacted information, accumulated information, apprenticed know-how; and talent and intuitive know-how. Finally, themes and issues which echoed those found in the Learning Organization Survey (Garvin et al., 2008) were considered in relation to the survey scores. For example, if a participant spoke about lack of training at the company, his or her response was considered along with the survey scores in the section of the survey dealing with training. Likewise, a recurrent theme in the group interviews dealt with the lack of information sharing; hence this recurrent theme was considered with the survey scores from the section on transfer and sharing of knowledge. In this manner, cross-validation was sought between the two research tools and the data they generated. In general, when analyzing the results, the theories associated with the learning organization concept and the knowledge-in-practice framework were used as templates for comparison and further theory development, rather than being imposed on the data, as recommended by Morse et al. (2002).

89

Chapter 4: Results Introduction This chapter analyses the results of the data collection. First, the results of the Learning Organization Survey (Garvin et al., 2008) are analyzed with a view to assessing the company’s strengths and weaknesses in order to provide a road map for future learning processes and initiatives in the study company. Then, the results of the semi-structured interviews are appraised, starting with the causal mapping method developed by Ambrosini and Bowman (2001); the relevant subsection focuses on key organizational constructs and success factors, their causes, and the knowledge required to realize them; it also includes analysis of numerous examples drawn from the interviews. In the subsection after that the semi-structured interview data are then analyzed using the knowledge-in-practice framework (McIver et al., 2013); in this subsection, degrees of tacitness and learnability of specific positions in the company are examined with a view to mapping these positions and key company activities to knowledge-in practice types. Survey Results The results from the Learning Organization Survey (Garvin et al., 2008) are examined in this section which begins with an analysis of the internal validity. This analysis is followed by a comparison of the medians provided by the authors to the company medians and means; the company means are then analyzed according to the benchmark quartiles provided by the authors. These two external comparisons allow for an assessment of the company’s strengths and weaknesses as a learning organization. Such an assessment can serve as a road map for the company to reinforce and build learning processes in the future, and to develop new initiatives such as training programs. Lastly, in this section, there is an internal comparison of survey

90

results across departments and levels of hierarchy. A table containing the complete survey results can be found in Appendix F; tables, including results according to department and levels of hierarchy, can be found in Appendix G. Internal validity. In order to check for internal consistency, the survey results were divided into two clusters according to date; then the averages, medians, and standard deviations were compared. The first cluster of survey results include the 22 surveys which were administered on the first three dates (December 2, 8, and 9) while the second cluster include the remaining 20 surveys which were administered on the remaining dates (December 14, 16, 18, and January 13). No significant differences between the clusters were noted which suggests that survey results were reliable over time. Furthermore, the means and medians were compared for the survey results across departments and across levels of hierarchy in the company. While there were some differences which are discussed in more detail below, overall there was a reasonable degree of similarity. The full results for this comparison can be found in Appendix G. Finally, distributions of the combined survey results for each building block were analyzed using five number summaries, box plots, and histograms all of which can be found in Appendix H. These distributions were further broken down and analyzed according to department and level of hierarchy. While there were some differences which are discussed in more detail below, overall there was a reasonable degree of similarity. External comparison. In this subsection, the survey results from the company are compared to the authors’ medians and benchmark scores.

91

Table 2. Comparison of medians provided by survey authors to company medians Building blocks and their subcomponents

Medians provided by authors

Medians from Difference company survey results

76 64 90 50 71

78.66 74.98 80.14 51.48 72.08

2.66 10.98 -9.86 1.48 1.08

71 80 71 80 71 74

64.26 71.40 72.91 49.98 52.79 62.87

-6.74 -8.60 1.91 -30.02 -18.21 -11.13

76

75.00

-1.00

Supportive Learning Environment

Psychological safety Appreciation of differences Openness to new ideas Time for reflection Learning environment composite Concrete Learning Processes and Practices

Experimentation Information collection Analysis Education and training Information transfer Learning processes composite Leadership That Reinforces Learning

Composite for this block

Table 3. Comparison of medians provided by survey authors to company means Building blocks and their subcomponents

Medians provided by authors

Means from company survey results

Difference

Supportive Learning Environment

Psychological safety Appreciation of differences Openness to new ideas Time for reflection Learning environment composite

76 64 90 50 71

76.99 73.08 80.16 52.27 70.62

0.99 9.08 -9.84 2.27 -0.38

71 80 71 80 71 74

62.61 66.81 74.28 50.72 50.66 60.61

-8.39 -13.19 3.28 -29.28 -20.34 -13.39

76

73.93

-2.07

Concrete Learning Processes and Practices

Experimentation Information collection Analysis Education and training Information transfer Learning processes composite Leadership That Reinforces Learning

Composite for this block

92

Table 4. Differences in median-to-median comparison, and median-to-mean comparison Building blocks and their subcomponents

Difference in medians

Difference in author’s median vs company mean

Supportive Learning Environment

Psychological safety Appreciation of differences Openness to new ideas Time for reflection Learning environment composite

2.66 10.98 -9.86 1.48 1.08

0.99 9.08 -9.84 2.27 -0.38

-6.74 -8.60 1.91 -30.02 -18.21 -11.13

-8.39 -13.19 3.28 -29.28 -20.34 -13.39

-1.00

-2.07

Concrete Learning Processes and Practices

Experimentation Information collection Analysis Education and training Information transfer Learning processes composite Leadership That Reinforces Learning

Composite for this block

Table 2 compares the survey authors’ medians to the company’s medians. Table 3 compares the same authors’ medians to the company’s means, and Table 4 extracts the differences from the two comparisons and presents them side by side. On the whole, the differences between the median-to-median comparison, and the median-to-mean comparison are quite close. For the first building block, Supportive Learning Environment, there are modest positive differences for Psychological Safety, and Time for Reflection; and there are slightly larger positive differences for Appreciation of Differences. There are negative differences, however, for Openness to New Ideas. The Learning Environment Composite shows a very small positive difference for the median and an even smaller negative difference for the mean comparison.

93

For the second building block, Concrete Learning Processes and Practices, all the differences are negative except for Analysis. The differences are greater than 10 for the last three items of this building block, and are particularly high for Education and Training, and Information Transfer. For the last building block, Leadership that Reinforces Learning, the differences are slightly negative. Table 5. Benchmark scores for the Learning Organization Survey (Garvin et al., 2008, p. 114) Scaled scores Bottom quartile

Second quartile

Third quartile

Top quartile

76 64 90 50 71

77-86 65-79 91-95 51-64 72-79

87-100 80-100 96-100 65-100 80-90

54-70 71-79 57-70 69-79 61-70 63-73

71 80 71 80 71 74

72-82 81-89 72-86 81-89 72-84 75-82

83-100 90-100 87-100 90-100 85-100 83-97

67-75

76

77-82

83-100

Building blocks and their subcomponents Supportive Learning Environment Psychological safety 31-66 Appreciation of differences 14-56 Openness to new ideas 38-80 Time for reflection 14-35 Learning environment composite 31-61 Concrete Learning Processes and Practices Experimentation 18-53 Information collection 23-70 Analysis 19-56 Education and training 26-68 Information transfer 34-60 Learning processes composite 31-62 Leadership That Reinforces Learning Composite for this block 33-66

Median

67-75 57-63 81-89 36-49 62-70

Table 5 presents the benchmark scores compiled by the authors of the survey—it reproduces the table which appears in the published article, “Is Yours a Learning Organization” (Garvin et al., 2008). The table includes quartile ranges and medians. According to the authors, it is the comparative scores which are more important than the absolute scores. Thanks to this table, the company means can be placed in quartiles; this allows for a more precise understanding of how the study company results compares to other company results. Some of

94

the authors’ quartile scores are surprisingly high; for example, the study company result for Openness to New Experience would have to be over 80 to surpass the first benchmark quartile. Table 6. Benchmark comparison for the combined company results for the Supportive Learning Environment building block

Supportive Learning Environment benchmark comparison with company results 120 100

The company Top Third Median Second Bottom

The company Top Third Median Second Bottom

The company Top Third Median Second Bottom

The company Top Third Median Second Bottom

The company Top Third Median Second Bottom

80

Pscyhological Safety

Appreciation of Differences

Openness to New Ideas

Time for Reflection

Learning Environment Composite

Bottom

66

56

80

35

61

Second

75

63

89

49

70

Median

76

64

90

50

71

Third

86

79

95

64

79

Top

100

100

100

100

90

76.99

73.08

80.16

52.27

70.62

60 40 20 0

The company

Table 6 presents the study company results in the last column (green) relative to the benchmark quartiles and medians for the building block, Supportive Learning Environment. Only the highest values of the quartile ranges are represented in this table. The study company survey results fall in the third quartiles for Psychological Safety, Appreciation of Differences, and Time for Reflection. The results are just barely over the bottom quartile for Openness to New Ideas, and they nearly tie for the median for the Learning Environment Composite.

95

Table 7. Benchmark comparison for the combined company results for the Concrete Learning Processes building block

Concrete Learning Processes benchmark comparison with company results

120 100 80

The company Top, 100 Third Median Second Bottom

The company Top, 97 Third Median Second Bottom

Experimentati Information on Collection

The company Top, 100 Third Median Second Bottom

0

The company Top, 100 Third Median Second Bottom

20

The company Top, 100 Third Median Second Bottom

40

The company Top, 100 Third Median Second Bottom

60

Analysis

Education & Training

Information Transfer

LPC

Bottom

53

70

56

68

60

62

Second

70

79

70

79

70

73

Median

71

80

71

80

71

74

Third

82

89

86

89

84

82

Top

100

100

100

100

100

97

62.61

66.81

74.28

50.72

50.66

60.61

The company

Table 7 presents the combined company results in the last column relative to the benchmark quartiles and medians for the second building block, Concrete Learning Processes. Once again, only the highest values for the quartile ranges are represented. All the items for the combined survey results are in the bottom quartiles except for Experimentation, which is in the second quartile, and Analysis, which is in the third quartile.

96

Table 8. Benchmark comparison for the combined company results for the Leadership that Reinforces Learning building block Leadership that Reinforces Learning benchmark comparison with company results 120 100 80 60 40 20 0

Series1

Bottom

Second

Median

Third

Top

The company

66

75

76

82

100

73.93

Table 8 presents the combined company results relative to the benchmark quartiles and medians for the third building block, Leadership that Reinforces Learning. Since there is only one set of questions for this building block, there is only one item on the chart. The combined company survey results fall in the second quartile. To sum up the external comparison, the company appears to be stronger overall in the first building block, Supportive Learning Environment, than the second building block, Concrete Learning Processes and Practices. The results for the third building block, Leadership that Reinforces Learning, are in between the first two. The results suggest the company’s biggest strength is in the item, Appreciation of Differences, because this item shows the most significant positive differences in the comparisons of medians-to-medians, and medians-to-means; furthermore, it falls in the third quartile in the benchmark comparison. Although not as large and only slightly above the authors’ medians, the items, Psychological Safety, and Analysis, also appear as potential strengths; the results for these items fall in the third quartiles in the

97

benchmark comparison. On the other hand, weaknesses appear in Openness to New Ideas, in the first building block; and Education and Training, and Information Transfer, in the second building block. These items not only show the most significant negative differences in the comparison of medians-to-medians, and medians-to-means, but the items also fall in the bottom quartiles in the benchmark comparison. However, the negative differences for Openness to New Ideas do not appear as significant as the negative results for the other two items because the absolute score for this item (80.16) is still quite high. The results provide evidence that the company has emergent learning organization characteristics, but is weak in some key areas, particularly in concrete practices and processes. These findings are discussed in more detail in Chapter 5. Internal comparison. This subsection analyzes the standard deviations for each item; it does so first for the whole company, second across departments, and third, across levels of hierarchy. High and low scores are also compared across departments and levels of hierarchy. Table 9. Standard deviation for the combined survey items. Highest standard deviations are highlighted Survey Item Psychological Safety Appreciation of Differences Openness to New Ideas Time for Reflection Learning Environment composite Experimentation Information Collection Analysis Education and Training Information Transfer Learning Processes Composite Leadership that Reinforces Learning

Standard Deviation 12.87 14.98 13.30 20.71 10.87 16.28 16.60 13.26 21.55 18.33 11.38 11.53 98

Table 9 shows the standard deviations for the combined study company results for each survey item. The highest standard deviations are found in Time for Reflection, Education and Training, and Information Transfer. Over all, the standard deviations appear higher for the second building block, Concrete Learning Processes and Practices. Table 10. Survey results across departments. Highest scores are in bold, lowest scores in red. Highest standard deviations are highlighted Department/Service

Operations N= 10

Commercial/ Marketplace N=9

Human E-commerce Resources N=16 N=5

Standard deviation

Psychological safety

80.04

75.91

86.32

72.2

6.05

Appreciation of differences

83.21

62.28

77.6

70.96

9.01

Openness to new ideas

87.09

77.36

81.41

76.77

4.76

Time for reflection

53.17

60.05

82.34

37.31

18.69

Learning environment composite

75.88

68.9

81.92

64.31

7.74

Experimentation

63.84

67.44

67.12

57.49

4.62

Information collection

64.97

73.78

67.12

68.28

3.75

Analysis

80.34

73.07

81.2

68.45

6.09

Education and training

44.74

55.8

79.02

40.46

17.27

Information transfer

40.45

59.44

66.91

44.86

12.37

Learning Processes Composite

58.87

65.91

72.27

55.91

7.34

Leadership that reinforces learning

83.5

71.94

80.5

67.66

7.36

Table 10 shows the survey results across departments including the standard deviations for each item. The finance department survey results have been removed because of the small sample size (N=2). The department which has the largest number of highest scores (seven out of eleven) is the human resources department; whereas the department with the largest number of lowest scores (nine out of eleven) is the e-commerce department. The items with the highest standard deviations across departments in Table 10 are the same items found in Table 9, which is to say Time for Reflection, Education and Training, and Information Transfer.

99

Table 11. Survey results across levels of hierarchy. Highest scores are in bold, lowest scores in red. Highest standard deviations are highlighted Department/Service

Executive Committee N=4

Managers N= 10

Employees N=28

Standard Deviation Across Levels of Hierarchy

Psychological safety

72.92

79.18

76.78

3.16

Appreciation of differences

62.48

78.92

72.51

8.29

Openness to new ideas

73.19

81.77

80.58

4.65

Time for reflection

49.34

54.05

52.05

2.37

Learning environment composite

64.48

73.48

70.48

4.58

Experimentation

66.94

64.62

61.27

2.85

Information collection

57.72

70.93

66.64

6.74

Analysis

77.20

76.34

73.12

2.15

Education and training

67.24

49.74

48.71

10.41

Information transfer

63.08

52.62

48.19

7.65

Learning Processes Composite

66.43

62.85

59.58

3.43

Leadership that reinforces learning

68.13

81.75

71.96

7.03

Table 11 presents survey results across the three levels of hierarchy of the company: the executive committee, managers, and employees. On the whole, there is less variation here than what is observed for the standard deviations across departments. The managers have the largest overall number of highest scores (seven out of eleven) whereas the executive committee has the largest overall number of lowest scores (seven out of eleven). In the second building block (Concrete Learning Processes and Practices), however, it is the executive committee which has the largest number of highest scores (five out of six), and the employees which have the largest number of lowest scores (five out of six). The highest standard deviation, 10.41, corresponds to Education and Training, which is also one of the items with the highest standard deviation found in the comparison across departments.

100

To summarize the internal comparison, an analysis of standard deviations for the company across items validates the results in the external comparison; both comparisons suggest the company’s main weaknesses lie in the second building block, Concrete Learning Processes and Practices. The larger standard deviations for these items indicate a wider range of responses and therefore less consensus among respondents. When standard deviations are calculated across the different departmental means, some of the per item standard deviations decrease; however, high standard deviations remain for Time for Reflection, Education and Training, and Information Transfer. Time for Reflection did not stand out in the external comparison because the mean score is above the authors’ median, and because it falls in the third quartile— nevertheless it is one of the lowest overall mean scores. While the sample sizes for the comparison of results across departments and levels of hierarchy are too small to meet the accepted criteria for statistical analysis of differences, they suggest divergence. This is particularly the case with the comparison across departments where recurrent significant per item differences can be observed between the human resources department, which tends to score high, and the e-commerce department, which tends to score low. While there is less variation when comparing standard deviations across levels of hierarchy, the Education and Training standard deviation stands out; as noted above, this item has the highest standard deviation in the overall company results, and the second to highest standard deviation across departments. Furthermore, the company mean score for Education and Training comes in second to last compared to all the other study company mean scores; also, the score shows the biggest negative difference when compared to the authors’ median score, and this score falls into the bottom benchmark quartile. Hence, there is strong and convergent

101

evidence that Education and Training is the company’s biggest weakness. These results are discussed in more detail in Chapter 5. Results from Semi-structured Interviews Causal mapping analysis. This subsection is organized according to organizational constructs and success factors which were identified during the initial interviews and/or during the individual interviews—these were refined throughout the data collection process in an iterative manner consistent with a qualitative research approach. Causal mapping was used to design the interview protocol in order to provide a loose structure to elicit tacit skills, and was applied to the analysis of the results as a framework for organizing the raw interview data. Diagrams map out the raw interview data starting with conceptual organizational constructs. These are, in turn, broken down into success factors which are more concrete expressions of the larger construct; causes, which are often enabling behaviors for the factors but sometimes include enabling external forces; and the knowledge required to achieve the factors. The diagrams are then explained and analyzed. Precise examples elicited by employees which illustrate the factors and enabling behaviors are described and commented. The boxes labelled “knowledge required” have been added to the causal mapping model which was described previously; this addition is an example of how the knowledge-in-practice framework enriches the causal mapping method.

102

Innovation

Factor A: anticipate changes and trends. Cause: inights, abilities, experience of founders and leaders. Knowledge required: experience with e-commerce, knowledge of the sector, creative thinking.

Factor B: specific company environ ment. Cause: founder's vision of managing a company. Knowledge required: how to transform visions and ideas into company policy: people management skills, understanding of labor laws and regulations.

Factor C: technical/product innovation.

Factor D: management innovation.

Cause: business decision , need for new business mode.l

Cause: management decisions/foun der's vision.

Knowledge required: knowledge of sector, technical know-how; how to encourage and manage technical innovation; knowledge of brands.

Knowledge required: creative thinking , abillity to convince others to try new ideas, ability to implement ideas.

Figure 6. Causal mapping of key organizational construct, innovation Figure 6 presents the organizational construct, innovation, and explores some of its expressions and manifestations in the company, as well as some of its knowledge pre-requisites and causes. Four factors include Factor A, anticipate changes and trends; Factor B, unique company culture; Factor C, technical product innovation; and Factor D, management innovation. Examples of Factors A and B were drawn primarily from the first individual interview with the co-founder and director of photography. Factors C and D were drawn from several different individual and group interviews. The interview data provides a number of different views about, and definitions of innovation found in the company, as well as providing some precise examples.

103

Factor A, anticipate changes and trends, results from the insights of the founders and leaders which were crucial to the initial success of the company. One of the most fundamental examples of this factor described by the co-founder is the original business model which involved offering a one-stop full-service e-commerce package to apparel brands who were previously not engaging in e-commerce at all, trying to manage e-commerce on their own, or hiring several different companies to take care of various aspects of their e-commerce. The founders identified a need for a more efficient service in an evolving sector and exploited this need. A second example, also described by the co-founder, was the merging of sales ecommerce sites with brand e-commerce sites. According to him, many brands were previously keeping these dimensions of e-commerce separate, and were not giving enough quality attention to the sites where consumers were actually choosing products and buying them. The more artistic and creative sites promoting the brand’s image and history were distinct from the shopping sites, which were largely text-based and less attractive visually. The founders anticipated the trend of combining these two functions—a trend which has now become the norm. A third example drawn from the interview with the co-founder involves his insights about the impact of smart phones on consumer behavior. According to him, smart phones are changing the way people shop for clothes, making them simultaneously more rational shoppers and more dependent on easily accessible smartphone-friendly images. Browsing in-person through brickand-mortar boutiques might lead shoppers to whimsical and spontaneous purchases, while smart phones encourage consumers to be more focused on function, onscreen images, and price comparison. Hence, the company is in the process of designing the e-commerce shopping experience accordingly.

104

Factor B, specific company environment, like Factor A, reflects highly tacit knowledge and insights stemming from the founders’ attitudes and values. Over time, the founders have developed and nurtured a very specific company environment. The company has moved several times and, at the time the interviews were conducted for this study, occupied three stories of a former parking garage which lent the facilities a loft like atmosphere—each floor was arranged into vast opens spaces. The founders wished to have no external signs of power differences so there are no offices and the dress code is very casual for everyone regardless of position. There is a collective refrigerator, which the managers make sure is well stocked at all times, and there are game rooms with sofas where employees can take breaks. The co-founder also explained a future project which is a company café where employees will be able to take their laptops and work whenever they want. Unlike the ability to anticipate trends, which exists primarily in the minds and words of the company’s leaders, these examples of the specific company environment are a more readily observable. While the company is currently in the process of rendering some its management practices into more formalized human resources policies, this is a very recent development. The practices observable today are largely the result of informal decisions, tacit process and routines that have developed over time. Factor C, technical/product innovation, is an approach to innovation which came later in the company’s history; experience with the original business model led the company’s leaders to re-evaluate some of their original assumptions and seek a different approach. Technical innovation was referred to briefly by the co-founder, and described in more detail by the commercial director as well as by some of the employees during the group interviews. The best example of this is a major research and development project currently underway to propose to brands a unique e-commerce platform tailored to the needs of the apparel sector. The company

105

plans to license the platform to its clients. This system will make use of artificial intelligence to better match potential client preferences to products being displayed on screen. Factor D, management innovation, explores ways in which the company implements innovative strategies in its daily management practices and operations. It is linked to the ability of the company’s leaders to anticipate trends and think creatively. One example of a very recent initiative was given by the human resources director; she referred to it as the “happiness equation”—an approach to employee well-being focusing on three areas: good use of time, quality of attention, and financial incentives. It is designed to introduce a more holistic approach to staff development. Another example of management innovation was described during the individual interview with the e-commerce manager who explained how the marketplace activity developed. The goal of this activity is to develop channels to help clients sell their products via online marketplaces. This activity was set up a year ago distinct from the e-commerce department because the company’s leaders judged it best to give the activity room to develop and grow. Now that it is doing well and generating turnover after only one year, it will soon be folded into the e-commerce department. A similar example was described about the recent creation of a new position and another new activity: the digital media manager. The goal of this position is to seek partnerships with various media providers and to leverage user-generated content on social media to boost sales and brands’ visibility. This activity is also being developed in a manner distinct from the rest of the company’s activities. These examples demonstrate that the company is able not only to anticipate trends but to set up functional positions and/or departments to exploit these trends. Such know-how is largely tacit and embedded in the company practices thanks to accumulated experience of what works and what doesn’t.

106

On the other hand, not all the examples about innovation provided during the interviews were positive. A negative example of Factor D came from one of the employees in the first group interview who stated that a position had been created in the past to be in charge of technological innovation but that this position did not last due to a lack of “tangible result”. It is not entirely clear whether the lack of result was due to the individual person, or due to the conception of the position—but the fact that the position was discontinued suggests the latter. This example illustrates a potential tension within the company between a long-term vision of innovation, which requires the willingness to experiment and invest, and the short-term expectation of a “tangible result”. This tension is a recurrent theme throughout many of the interviews. It may also illustrate a failed approach to innovation which relied too much on an individual rather than on more collective processes—this example is discussed further in Chapter 5 in the Theory subsection.

107

Figure 7. Causal mapping of key organizational construct: commercial success

Commercial success

Factor A: earning the client's trust.

Cause: well-defined process for assessing if the client is a good fit and tailoring the offer.

Knowledge required: how to analyze the client's e-business using a rubric; how to understand the client's needs, how to listen.

Factor B: getting new clients.

Cause: making an attractive offer/ generating word of mouth.

Knowledge required: how to adapt and think creatively; knowing exactly what the company can offer to the client and what the costs are; how to convince the client.

Factor C: turn costs into profit.

Cause: "clarifying the perimeters" .

Knowledge required: how to analyze the company's needs using a rubric; how to identify which services are crucial to the client and which services should be offered as "extras".

Figure 7 presents the key organizational construct, commercial success, which is then broken down into more concrete factors including Factor A, earning the client’s trust; Factor B, getting new clients; and Factor C, turning costs into profits. Figure 7 is largely drawn from the third individual interview which was with the head of commercial development and strategy; he

108

has been in this position for one year and a half. A few of the examples of the factors discussed below, however, come from other individual and/or group interviews. Factor A, earning the client’s trust, is caused (or enabled) by well-defined processes to assess the client’s needs; it requires a mix of communication skills and analytical skills. One example of Factor A was provided by the head of commercial development and strategy who described how his team had recently managed to earn a client’s trust despite initial resistance. With the aid of a rubric specially developed for this type of analysis, the commercial team assessed how much the client was spending on its current e-commerce activity and the team was able to redesign a package of services which would save the client money in the long run. A negative example of Factor A, however, was given by the co-founder who spoke about a client which the company had lost because they had not been able to explain to the client’s directors how to sell their products via e-stores, and because they could not agree on objectives. In the negative example, it is unsuccessful communication and interactions rather than lack of convincing or effective analysis which is responsible for the loss. Factor B, getting a new client, is caused (or enabled) by making an attractive offer and/or generating word of mouth. This factor requires a combination of knowledge and skills including highly analytical skills, listening skills, and the ability to adapt and think creatively. The example given by the head of commercial development is somewhat similar to the example given for Factor A (earning a client’s trust). It consists of a story he related about how, after an initial offer to a client was refused, the commercial team made a new offer focusing only on traffic analysis and ad words. In the new offer, the company would earn commission as a percentage of return on investment; hence the company would only make money if the client’s traffic and sales increased thanks to their tailored e-commerce services. The offer was accepted

109

and the head of commercial development believes they will be able to expand the offer later to include other services. Factor C, turn costs into profits, continues with the idea that by a rationalization of their activities, the company can improve their bottom line. The cause (enabling behavior) for this factor is described as “clarifying the perimeters” which refers again to the new strategy of making tailored as opposed to full-service offers. Such an approach requires a discriminating ability to identify what is crucial to the services and expertise offered as opposed to what is “extra”. For example, the after sales and the logistics services the company includes in its full service package would be offered only as extras—fees would be calculated accordingly. This fee structure potentially transforms these extra services into revenue earning services instead of costs.

110

Good relations with the clients

Factor A: maintaing good relations with the client interlocutors.

Cause: employees' abilities to communicate effectively with the client interlocutors.

Knowledge required: communication skills especially listening skills, knowledge of the brand, ability to explain technical problems to the client, abiliy to talk to technical emplooyees.

Factor B: client satisfaction with ecommerce websites (delivery, quality, maintenance).

Cause: ability of ecommerce and technical employees to understand clients' demands, cooperate, and deliver the websites on time.

Knowledge required: how to train new employees, ability to anticipate, demand, manage work flow, communicate across departments.

Factor C: effective internal communications and cooperation. Cause: flow of information across teams and departments; productive cooperative attitude of employees.

Knowledge required: organization-level communication skills, team work skills, people skills .

Figure 8. Causal mapping of key organizational success factor: good relations with the clients

111

Figure 8 presents the key conceptual organizational success factor, good relations with the clients, which was a recurrent theme throughout the interviews. This construct is broken down into the following factors: Factor A, good relations with the client interlocutors; Factor B, client satisfaction with the e-commerce websites; and Factor C, effective internal communications and cooperation. Figure 7, commercial success, and Figure 8, good relations with clients, are sequentially linked in many ways because, after the company gains a new client, it has to keep the client—maintaining good relations is crucial for this. Moreover, the two figures, and the examples discussed after each one, are complementary, because the raw data providing the content for most of Figure 8 come from the group interviews with employees who work in e-commerce and operations; these data provide a different perspective of client relations than those provided by the co-founder or the commercial director and head of strategy. Figure 8 represents the view from the employees who have to provide the services promised to the client by the sales team after the deal is signed. Factor A, good relations with client interlocutors, is key because the relationship between the company and its client is heavily influenced by the type of relationship between individual representatives from each side. Company employees must deploy effective communication skills to maintain as productive and smooth a relationship as possible with whomever is their client interlocutor. Furthermore, they need a firm technical and operational understanding of ecommerce to be credible interlocutors. An example of this was discussed in Group Interview 3 when e-store employees talked about the challenge of having to diplomatically refuse client requests that were unreasonable, infeasible, or simply not good for e-commerce. Employees facing this delicate task have to be charismatic while also demonstrating that their views are

112

based on solid expertise and experience. This example is developed more below in the context of the key organizational construct, employee and client development. Factor B, client satisfaction with e-commerce sites, goes to the core the company’s competencies. It requires a complex bundle of skills and knowledge which cut across various departments and positions including the ability to manage work flows, and train new employees. One of the key causes (enabling factors) is the ability of the technical employees to meet demand. A negative example of Factor B was provided by an employee who described a crises period during which many key employees left the company just when the sales team had brought in several new clients. This unfortunately coincided with a period when key technical people had been assigned to work on developing the new platform; these employees were thus not available to work on e-commerce sites. Not surprisingly, this situation led to client dissatisfaction as e-commerce sites, or features of sites, or site improvements were not delivered on time and/or were delivered with technical problems. A second negative example was explained by another employee who described what happened when a former colleague, who knew a great deal about setting up payment systems, suddenly left the company. Because there had been no effective transmission of his knowledge, nor an effective rendering of it in any explicit or retrievable form, the knowledge was lost. The company was not able to set up or modify a crucial feature of e-commerce sites in a timely and efficient manner. This example illustrates how staff turnover can negatively impact the company’s ability to provide an essential service when the knowledge is not effectively stored and accessible to other employees. A third negative example was provided by a technical employee who stated that too much of his time was taken up by having to explain to other

113

employees how to accomplish basic technical tasks because they hadn’t been properly trained; this prevented him from efficiently progressing on other projects. Factor C, effective internal communications and cooperation, is caused (or enabled) by the flow of information across teams and departments, and by the productive cooperative attitude of employees. This factor really undergirds almost all the other organizational success factors, and it requires organizational competencies and team competencies in addition to individual ones. One employee, for example, described a situation when there were many interns working for the company who were not informed about and/or did not have access to key pieces of information. This led to intermittent and unreliable communication with the client especially when it came to explaining difficulties the company was facing delivering the client’s ecommerce site on time. The employee furthermore asserted that the client interlocutor would not necessarily have been so unhappy if the company had simply informed him in advance that they would not be able to meet the deadline. Another example of Factor C was provided by an employee who deals with digital growth; she expressed frustration with what she felt was a highly centralized and poorly communicated commercial strategy. In her view, this harmed her ability to cultivate good relations with the client interlocutors because she was not knowledgeable enough about her own company’s strategy. Both of these examples not only illustrate Factor C but also demonstrate how intertwined factors A and C are. An additional example of Factor C was provided by a technical employee who described what, in his view, was a key skill for effective team communication and cooperation. He used the phrase “lacher-prise”, which literally means letting go. He used this phrase as a way to explain that employees should not be individually too attached to one specific solution or approach to a problem but should be ready to try something new if the original approach doesn’t

114

work. Given the complex mix of business, marketing and technical skills required for realizing e-commerce sites, no single employee is likely to have a monopoly on the relevant knowledge and expertise. While flexibility may be considered individual trait, it can also be considered a skill when it can be developed over time, with practice, and when it is linked to performance (Matteson, Anderson, & Boyden, 2016).

Employee and client development (faire grandir)

Factor A: employee autonomy (learning about themselves, taking risks, gaining skills).

Factor B: client autonomy (able to determine objectives, assume responsibility, anticipate).

Cause: managers create opportunities for employees to learn and increase their sense of autonomy.

Knowledge required: people management skills, how to create learning opportunties.

Cause: the company educates, informs, trains the client.

Knowledge required: knowledge of e-commerce , of sector, and specific knowledge of the client's priorities and corporate culture, communication skills.

Figure 9. Causal mapping or organizational success factor: client and employee development Figure 9 presents the conceptual organizational success factor, employee and client development, which was the best way of translating a recurrent phrase in the interviews: faire grandir. Literally, the phrase means to make grow but for the study company if refers to developing a sense of autonomy internally, i.e. among employees (Factor A), and externally, i.e.

115

among clients (Factor B). The knowledge required for this is primarily management and people skills. Managers need to be skilled at creating opportunities for employees to take the initiative and learn; while, with the client, the company representatives need to be skilled at transferring some of their knowledge and skills to the client interlocutors to achieve a smoother and more productive working relationship. One example of Factor A, employee autonomy, was given by the human resources director who explained that she had delegated responsibility of choosing a supplier to the employees who would form teams and attend presentations by supplier representatives to make the final choice. She further explained that the company would be moving facilities in a matter of months and that employee input was actively solicited for the design and furnishings of the new space. The human resources director stated that employees at the company were expected to “dare” to make decisions. A second example of Factor A was given by an employee who worked in the anti-fraud section of the finance department and who had created an internal newsletter to educate other employees about fraudulent transactions and their significance. This initiative has the potential to stimulate a virtuous circle by empowering other employees with knowledge about fraud who may then take the initiative themselves to modify their behavior in ways that help reduce the likelihood of fraudulent transactions. A third example of Factor A, employee autonomy, was given by a brand manager who explained that she regularly asks the e-store managers who report to her key questions about their activity—questions about their best selling products, the product display on the websites, and the investments in media. These questions are meant to stimulate the e-store managers to think more critically and independently about what they do on a daily basis. The same brand manager also stated that she expects her e-store managers to occasionally challenge the advice

116

from the traffic department because this demonstrates their autonomy and confidence in their own judgment. On the other hand, a negative example of Factor A came from a technical employee who had conducted research on his own time about source codes to be used on servers; he had wanted to deploy a new source code but encountered resistance on the part of his manager. According to him, employees are not encouraged to take the initiative, and innovation in the company is rare; furthermore, suggestions are not taken seriously by managers unless they come from the clients. Another similar negative example came from an employee who explained that he had been made to redo a checkout system three times without any real justification and without achieving a better result. According to him, many such futile initiatives are undertaken simply because managers have to “prove they’re right”. These examples suggest that the positive view of employee autonomy and development is not shared by all. Furthermore, the difference between the positive and the negative examples suggest that tacit people management skills have been developed effectively by some managers more than others. Regarding Factor B, client autonomy, several examples and/or statements from employees illustrate inherent challenges. One such example illustrates concretely what happens when the client doesn’t take sufficient responsibility for the e-commerce activity. An employee described a situation in which the client did not anticipate ordering enough stock for an upcoming sale. Hence, the efforts on the part of the company to boost traffic and promote products were useless because the client could not provide the products to the customers. An additional example of Factor B came from an e-store manager who frequently has to deal with client requests that he feels are unreasonable, unfeasible, or counterproductive. In such cases, he considers that his role is to say “no” to the client in a diplomatic enough way to avoid

117

jeopardizing the relationship. This requires delicate communication skills, charisma, and persuasion. He stressed how important it is to convince the client in such moments—backing up arguments with hard evidence from running tests, for instance, is sometimes necessary. Finally, another interesting example which illustrates Factor B came from an employee who was embedded in the client company for a temporary mission—this position is described and analyzed in more detail in the following subsection Knowledge-in-practice framework analysis. Her experience working directly with interlocutors from the client for several months led her to suggest that the study company think more about what e-commerce knowledge needs to be shared with the client representatives so the clients gain a more complete understanding of the benefits and challenges of the activity. Ultimately, the goal of client development is to educate the clients so they can better assess independently the impact of their own behavior and decisions on their e-commerce activity and strategy. Individual positions were also analyzed using the causal mapping methodology—the maps for these positions can be found in Appendix I. A few tacit skills not captured in the maps above dealing with organizational constructs were revealed thanks to the mapping of these individual positions. For example, the positions of brand manager, and digital media manager both require the ability to work in a dynamic, fluid context with loosely defined processes; also, they require the ability to prioritize under stress, and to deal with uncertainty. One interesting example of a situation requiring such skills came from a brand manager whose client brand had recently been acquired by a Turkish company. This new company new so little about their own products, they were unable to provide even the most basic information about their stocks. The brand manager thus had to personally visit warehouses and devise methods to estimate the quantity and value of the client’s stock. Another relevant example came from a digital media

118

manager who explained that the study company’s largest client had just been promised an advertising package which would substantially increase traffic and revenues; however, the costs of the advertising package had not been properly evaluated—she therefore found herself in the delicate position of discovering that the costs would likely far outweigh the estimated impact on revenues. To sum up, the organizational construct, innovation, is associated with highly tacit skills particularly when it comes to anticipating changes and trends. Furthermore, the specific company environment has formed over time thanks to the vision, and accumulated experience of the founders; this vision and experience have become embedded in the company’s tacit routines and practices; the unique environment is observable in the company floor plan, dress code, breakroom, etc. This environment is highly related to management innovation, where we find approaches and methods developed over time for creating new positions and exploiting potential opportunities; these approaches are unlikely to be written down in any manual. Meanwhile, the recent orientation towards technical innovation requires the company to render its accumulated tacit knowledge gained through experience into a more explicit vehicle—an e-commerce platform specifically for the apparel sector. The organizational construct, commercial success, appears to be based less on tacit skills than the innovation construct because there is more emphasis on explicit analytical skills and tools. Nevertheless, there is still the need for tacit communication skills when it comes to explaining to the client the potential benefits of the company’s e-commerce offer. While analytical rubrics are indispensable, the commercial team’s ability to convince would be most effective when combined with the accumulated experience of the company assimilated and

119

disseminated largely through tacit exchanges among members of the team and with other employees. The organizational construct, good relations with clients, requires organization-level skills which encourage the effective flow of information across different teams and departments; furthermore, it requires organizational-level skills to manage work flows, anticipate demand, and train new employees. This is tacit to the extent that employees engage in productive behavior without thinking about what they’re doing, and without it being formalized. This appears to be the case but the behaviors are uneven and unreliable; hence more formalized processes are most likely necessary here. Maintaining good relations with clients also requires individual-level tacit skills such as the ability to communicate well with the client. Finally, effective team work necessitates individual-level tacit soft skills such as flexibility, and the ability to share responsibility for outcomes. The organizational construct, employee and client development, requires people management skills in order to create opportunities for employee initiative and learning. Many of these skills are tacit such as the ability and/or willingness to delegate, and the ability to ask employees the right questions. Once again, these practices appear to be uneven, and unreliable. This area would therefore also most likely benefit from a more explicit and formalized approach. Client development meanwhile requires a mix of tacit skills such as communication skills, the ability to understand the client’s point of view, and knowing what kind of information can and should be shared. Other tacit skills that were revealed pertaining to individual positions such as brand manager, and digital media manager include the ability to work in a dynamic, fluid context with

120

loosely defined processes, as well as the ability to prioritize under stress and to deal with uncertainty. Knowledge-in-practice framework analysis. The knowledge-in-practice framework, described in Chapter 3, includes four different knowledge-in-practice (KIP) types which are differentiated by their respective degrees of tacitness and learnability. These practice types are represented once again below in Figure 9 but in a slightly different visual form than what was presented in Chapter 3. Figure 9 makes it clearer how the KIP types relate to each other according to the varying degrees of tacitness and learnability: High learnability Low tacitness, high learnability

High, high

Low tacitness

High tacitness

High tacitness, low learnability

Low, low

Low learnability

Figure 10. Knowledge-in-practice framework (adapted from McIver et al., 2013, p. 601)

121

In this section, a few key activities/positions which best represent the knowledge-inpractice types in the company are analyzed using the framework. A more exhaustive analysis of the positions in the company can be found in Appendix J. Activity/Position

What is high performance?

After sales service/ Customer service

Quantitative: number of calls and e-mails sent. Qualitative: good understanding of the brands, ability to talk to customers and resolve problems and potential conflicts.

How is high performance achieved? Combination of clear written processes, and procedures, solid stable organization and routines with team building, guidance, and support.

What needs to be known? Effective and efficient communication skills (written and oral): ability to answer calls, talk to clients, respond to email inquiries rapidly and accurately. Knowledge of brands: their specific characteristics, processes, issues affecting delivery, etc.

How does knowing take place? Introductory training, access to stored information through manuals and diagrams. On-the-job training with constant feedback with managers and colleagues.

How is knowing applied?

KIP type

Ideal work design will involve replication and reuse of routines, storage and easy access to work practice information, continuous process and outcome based performance feedback.

Enacted information (low tacitness, high learnability) with elements of apprenticed know-how (high tacitness, high learnability) introduced by the customer service manager who stresses oral transmission, coaching, and a team dynamic.

Figure 11. Knowledge-in-practice analyses of after sales service activity Figure 11 presents an analysis of the after sales service activity. There are many positions in the company which require a high-level of technical knowledge, specialized expertise, and/or an understanding of the business. As explained in the Chapter 2 subsection Past, present, and future of e-commerce, e-commerce has evolved at a rapid rate since it began in the late 1990s, and it continues to do so in ways that are immensely challenging for companies to keep pace with. Hence it is not surprising to discover there are few activities in the company which are stable, highly repetitive, and relatively easy to learn for someone lacking higher studies and/or specialized training. Therefore, few positions in the company correspond to the knowledge-inpractice type referred to as enacted information, which McIver et al. (2013) characterize as being low in tacitness and high in learnability. According to McIver et al., this is typical of activities such as grocery cashiers, bank tellers, waiters, and manufacturing jobs. As demonstrated in Figure 11, the after sales service activity, corresponds to this knowledge-in-practice type. This

122

department has the lowest degree of interface with other departments in the company; moreover, it can easily be identified visually in the open space floor plan because it occupies a distinct zone where the employees face each other in mini cubicles—this appears to be a practical necessity since these employees spend a significant amount of their time answering calls and need to be located in a zone without excessive background noise. The current head of this department trains new employees by using both explicit, stored information (manuals and diagrams), and feedback/coaching. He furthermore strives to develop a team dynamic among the customer service employees by encouraging feedback and knowledge-transfer among the staff, and has even introduced a modest element of self-leadership by allowing employees to set goals for themselves, and to request working with specific brands if they develop preferences during their training period. Moreover, the information about specific brands (their products, their delivery times, etc.) which must be assimilated to effectively carry out this work, is broad and complex. Hence, there are facets of this activity which do not correspond perfectly to the enacted information practice type. Thanks to the leadership of the after sales service manager, and because of the complexity of the information involved, there are elements of apprenticed know-how, particularly in the opportunities the employees have to hone and refine their work. This augments the otherwise low tacit dimension of the activity, particularly for the after sales service manager because of the amount of time and energy he spends giving feedback and coaching customer service staff on the job.

123

Activity

E-store manager (senior)

What is high performance? Selling the right product at the right time, breaking sales records, delivering the site or modification to the client on time.

How is high performance achieved? Targeted promotional initiatives such as newsletters. Effective technical and operational knowledge and support which is built on routines and processes, availability of information about products, and logistics.

What needs to be known? Understanding of the product and a business sense for selling products. A sense of good timing for sales and promotions, which products to feature and display on line. Sufficient technical knowledge to talk to technical people, and understand and anticipate technical constraints. How to communicate effectively with clients. How to work effectively with junior e-store manager.

How does knowing take place? Business sense has to be honed and cultivated through experience and being “open to learning from the market”. Technical knowledge and skills can be transmitted through written processes and procedures.

How is knowing applied?

KIP type

Ideal work design would build on routines and work settings that facilitate storage and access to work practice information while providing opportunities for mentorship and incident –triggered outcomes-based performance feedback. Furthermore, there would be systematic but flexible ways to transfer and share expertise.

This is a layered position combining aspects of enacted information (low tacitness, high learnability), and accumulated information (low tacitness, low learnability). It builds on the former, which corresponds to the more technical (junior) aspects of the position in order to facilitate the latter, the business analysis aspect.

Figure 12. Knowledge-in-practice analyses for senior e-store managers Figure 12 presents the senior e-store management activity using the knowledge-inpractice framework. E-store managers work in pairs; the junior e-store manager deals with the more routine and operational facets of the e-store activity, which best fit the characteristics of enacted information, because there is a significant degree of stable routines when it comes to making e-commerce sites operational, updating them, and handling operational concerns such as deliveries. The senior e-store manager meanwhile deals more with the analysis of business cycles, consumer tastes and behavior, marketing, and communications; this activity therefore best fits the characteristics of accumulated information. While the two positions are analyzed separately using the KIP framework in Appendix J to clarify their differences, in reality, there is no clean separation between them because roles and responsibilities overlap; each pair of e-store managers works together differently according to the temperaments and skills of the two individuals. One important tacit skill inherent in both the junior and senior e-manager positions would be the ability to share responsibilities and communicate effectively with the other e-store manager. 124

Activity

Lead architect (in charge of IT infrastructure)

What is high performance? Successful realization of projects, improvements, updates

How is high performance achieved? Application of extensive and complex expertise through the use of logical and systematic analysis.

What needs to be known? Key cycles and dates for brands, understanding of technical constraints; how to manage others. Extensive, complicated and detailed information that must be understood in relation to contingent factors for situated analysis.

How does knowing take place? Extensive study and prior experience to master specialized expertise, numerous systematic and analytic search processes and problem solving accompanied by continuous learning, adaptations, and synthesis of new information.

How is knowing applied?

KIP type

Ideal work design would encourage the application of specialized expertise, systematic but flexible ways to share and transfer this expertise, opportunities for education and reflection, outcomes and process-based performance feedback.

Accumulated information (low tacitness, low learnability)

Figure 13. Knowledge-in-practice analyses for lead IT architect, and the supply chain logistics manager Figure 13 presents the lead IT architect activity, a position in the operations department. Other activities from this department, including the supply chain logistics manager, and the product manager (Saas), are analyzed in Appendix J. These activities best fit characteristics of accumulated information because they are relatively low in tacitness and learnability. They all require people to have done extensive study, and to have acquired prior experience in order to master the complex knowledge and expertise necessary to design customized solutions and to deal with unexpected parameters.

125

Activity

What is high performance?

An e-commerce position embedded in the client company.

Optimal balance between the client’s ecommerce autonomy and the company’s support and guidance.

How is high performance achieved? A near perfect sharing of information so that each side can fulfill its role.

What needs to be known? Broad scope of knowledge about the client brand, its corporate culture and practices. Deep understanding of e-commerce, the services provided by the company, and the practices and culture of the company.

How does knowing take place? Total immersion in the client company to assimilate its culture and practices. Learning by doing (ecommerce platforms). Exchanging and proactively engaging with interlocutors (both client and company).

How is knowing applied?

KIP type

Ideal work design would encourage the application of specialized expertise, systematic but flexible ways to share and transfer this expertise, opportunities for education and reflection, outcomes and process-based performance feedback.

This position is partly accumulated information (low tacitness, low learnability) – the person functions like a consultant, an auditor or analyst working in the client firm. But there is a high degree of adaptability required (hence tacit skills). There are elements of apprenticed know-how, and talent & intuitive knowhow.

Figure 14. Knowledge-in-practice analyses for an embedded e-commerce position Figure 14 presents the activity of an e-commerce position which is embedded in a client company. The person working in this position participated in the fourth group interview—she had replaced a digital assistant out on maternity leave who was employed directly by the client company. The role of this position is similar to a consultant’s role or a financial analyst when such people carry out missions for defined, temporary periods in client companies, generally to accomplish specific projects, or to conduct audits. Such activities are usually characterized as accumulated information because they are low in tacitness and learnability (McIver et al., 2013). The embedded position presented above, however, does not seem to be low in tactiness. Hence, it provides an interesting example first because it does not fit neatly into the KIP typology and second because the position was described during the interview with the e-commerce manager during which it led to a discussion about his vision of the company as a hub. This vision is one in which the company primarily provides and shares expertise—employees spend time in client firms, and vice versa. This position does require a high degree of e-commerce expertise, not to 126

mention specialized knowledge about how the study company functions (hence accumulated knowledge); but the person would also have to be highly skilled at learning on the job while immersed in the client firm in order to quickly assimilate the firm’s practices and to be able to function effectively as a bridge between client and company. Furthermore, the daily job of this position could vary greatly from one client to another. Consequently, tacit skills are significant because the person would have to be an astute observer, able to learn quickly from exchanges with colleagues, and be highly adaptable. Therefore, in addition to elements of the accumulated Information type, there are significant elements of apprenticed know-how due to the crucial importance of feedback and honing skills. At first view, this appears contradictory because apprenticed know-how is the opposite of accumulated information in terms of both tacitness and learnability. But on-the-job learning, trial and error, honing techniques are necessary to effectively apply accumulated information and specialized expertise to daily problems which constantly change.

127

Activity

What is high performance?

Brand Manager (ecommerce)

Successful positioning, visibility and development of the brands.

How is high performance achieved? Anticipating trends and creatively thinking about how the brands can develop, maintaining good relations with the client, developing the teams working with the brands.

What needs to be known? A global understanding of the “brand’s ecosystem”, a deep understanding of e-commerce and ecommerce platforms, a strategic business sense of what drives traffic and turnover, how to communicate with the client, people management and team building skills

How does knowing take place? A brand manager must have learned and assimilated robust technical and business knowledge through study and past experience. On top of that, brand managers must learn on the job – feedback and mentoring play a significant role.

How is knowing applied?

KIP type

Ideally work would be designed to facilitate the application of the specialized expertise with ways to share and transfer knowledge with colleagues. But there should also be an emphasis on application of experience based skills and developing those skills through related experienced. Apprentice/mentor relationships should have an important role and constant outcome-based feedback. There should be new challenges and opportunities to develop selfleadership.

Hybrid between Accumulated Information (low tacitness and low learnability) and Apprenticed Know-how (high tacitness and high learnability) with dimensions of Talent and Intuitive Knowhow (high tacitness, low learnability)

Figure15. Knowledge-in-practice analysis for e-commerce activity: head of traffic and CRM, and brand manager positions Figure 15 presents the activity of the brand manager. This activity fits the characteristics for the accumulated information KIP type when it comes to the extensive knowledge and expertise required, but, like the embedded e-commerce position described above, it does not fit this type perfectly. Because it involves a significant degree of learning on the job, and constant outcome-based feedback to continuously improve and learn, some elements are a bitter fit with the characteristics of the apprenticed know-how KIP type. Meanwhile, a phrase such as “the brand’s ecosystem”, elicited by a brand manager during the Group Interview 5, emphasizes the importance of a global understanding of e-commerce and the ability to make links across its diverse components. The ability to make good decisions based on this global understanding is a complex, ambiguous, hard to explain, and constantly evolving skill. Hence, there are dimensions of talent and intuitive know-how in both of these positions, which is to say high tacitness, and

128

low learnability. The work practices associated with the brand manager position therefore vary greatly—they may involve knowledge and skills which range from being highly tacit to barely tacit, and likewise from being relatively easy to learn to extremely hard to learn. In this way it is similar to the position of head of traffic and customer relations management which is also analyzed using the framework in Appendix J. Activity

Co-founder/ Head of Photography

What is high performance? Ability to anticipate change and offer new types of services/products needed by the evolving market

How is high performance achieved? Innovative ideas, willingness to take risks, ability to reach and convince key decision makers for brands in the sector.

What needs to be known? Evolving understanding of e-commerce trends in the sector, broad and constantly evolving understanding of how the Internet functions and how to innovate on the Internet, knowledge of the potential and limits of technical innovations, knowledge of brands (their products, needs, and customers),

How does knowing take place? Studies and experience are required to gain the base of knowledge in ecommerce. Ideal prior experience in brand management (or something closely related). On the job, knowing takes place through conditions that nurture innovation, creativity, aligning diverse efforts, exploration, and experimentation.

How is knowing applied?

KIP type

Ideal work design leverages unique talent and abilities, initiates new challenges. It encourages self-leadership and social network development.

Talent and Intuitive know-how (high tacitness, low learnability)

Figure 16. Knowledge-in-practice analysis for co-founder/head of photography In Figure 16, the activity of the co-founder/head of photography is presented—the data comes primarily from the first individual interview. This co-founder, while today less involved in the daily commercial operations of the firm than he was a year and a half ago, provided information about the history and development of the company which allows for an extrapolation of the activity of all three founders of the company. It provides the clearest example of the talent and intuitive know-how KIP type (high tacitness, low learnability) because of the emphasis on innovation, exceptional outcomes, leveraging invention, aligning diverse efforts and selfleadership. The kind of innovation and anticipation referred to by the co-founder during the

129

interview is difficult to teach because it draws on intuition and inspiration. The highly tacit skills involved are complex, ambiguous and constantly evolving. High learnability

Low tacitness, high learnability

High, high

Customer Service

Low tacitness

E-store manager

High tacitness

Embedded position Brand manager

Co-founder/Head of Photography

Low, low Low learnability

High tacitness, low learnability

Figure 17. Knowledge-in-practice diagram with key job positions (adapted from McIver et al., 2013, p. 601) Figure 17 reproduces the diagram from Figure 10 but adds a mapping of the key positions which have been analyzed in Figures 11-15. This allows for a dynamic visualization of the positions on continuums from high to low tacitness, and from high to low learnability. Because of the importance of specialized, technical skills and knowledge in the company, the quadrant where most of the positions cluster together is the accumulated information KIP type. Almost all of the positions analyzed using the framework above and those in Appendix J have elements

130

which fit the characteristics of this KIP type. The customer service activity, and the co-founder, which represent two opposite poles, come closest to representing pure types. According to the framework then, we would expect the customer service activity to be relatively easy to learn and to rely mainly on explicit knowledge and skills, whereas the latter would be very difficult to learn and would rely mainly on tacit knowledge and skills. The other positions in the company would be somewhere in between.

High learnability Low tacitness, high learnability,

High, high Commercial and business analysis; aspects of various activities which involve oral transmission, feedback, coaching, and mentoring, and learning through doing.

Dealing with incoming final consumer questions and requests; dealing with incoming messages from service providers and company clients; routine, technical aspects of setting up sites and managing deliveries.

Low tacitness

Low, low

High tacitness

Finding customized solutions to clients’ needs and problems; prioritizing and problem solving under pressure. Analysis of traffic and SEO; maintaining IT infrastructure; managing logistics

Creative conception of websites; developing and communicating a vision of an ecommerce ecosystem; anticipating ecommerce trends; thinking strategically; risk taking, experimentation.

Low learnability

High tacitness, low learnability,

Figure 18. Knowledge-in-practice diagram with key activities (adapted from McIver et al., 2013, p. 601)

131

Figure 18 also reproduces the diagram from Figure 10 but this time presents the main activities—rather than positions—of the company according to the KIP types they best fit. Figure 18 provides a useful complement to Figure 17 because several positions involve activities which straddle more than one KIP type, and because there is considerable flexibility in the company’s management culture. Employees are not necessarily expected to limit themselves to rigidly delineated job descriptions and responsibilities. The ability to adapt to circumstances, and to take initiatives is highly valued, not to mention to willingness and ability to share different points of view at all levels of the hierarchy. To sum up, two positions analyzed using the knowledge-in-practice framework (McIver et al., 2013) appear as straightforward matches to the KIP types: the customer service activity fits the characteristics of enacted information because it is low in tacitness and high in learnability; and the co-founder/head of photography position fits the characteristics of talent and intuitive know-how because it is high in tacitness and low in learnability. The other positions analyzed using the framework are clustered around the middle of the diagram. While they fit many of the characteristics of accumulated information (low tactiness, low learnability), none of them are perfect matches. They all straddle two or more different KIP types. For example, the position of e-store manager has routine and repetitive aspects to it as well as more complex dimensions which are harder to learn and require the subtle application of expertise; hence the e-store manager activity is both enacted information (low tacitness, high learnability) and accumulated information (low tacitness, low learnability). The embedded e-commerce position requires mastery of complex information as well as on-the-job learning and continuous adaptation and honing of skills based on feedback; hence the activity fits the characteristics both accumulated information (low, low) and apprenticed know-how (high, high). The brand manager is the most

132

complex position because it requires the mastery of complex information, on-the-job learning, continuous adaptation and honing of skills, as well as creative thinking, business analysis, and intuition; hence it combines dimensions of accumulated learning (low, low), apprenticed knowhow (high, high), and talent and intuitive know-how (high tacitness, low learnability). When separated from specific positions, however, the activities of the company can be placed easily in each of the four KIP types. Repetitive tasks such as processing final consumer requests, deliveries, or dealing with routine technical aspects of setting up websites are low tacit and highly learnable activities, hence enacted Information. Business and commercial analysis, which requires understanding and anticipating business cycles and consumer behavior, or any activity which is heavily dependent on feedback, honing skills, and requires continuous improvement, is highly tacit and highly learnable on the job, hence apprenticed know-how. Complex problem solving and the maintenance of systems which require the mastery of large quantities of specialized information and skills are low tacit and difficult to learn activities, hence accumulated Information. Finally, the more creative, strategic, and risk taking activities at the company correspond to the talent and intuitive know-how type. Answers to Research Questions This section examines each of the research questions noted in Chapters 1 and 3. Research question a: How are the concepts of organizational learning, knowledge management, and the learning organization related? In Chapter 2, I surveyed the literature on organizational learning, the learning organization, and knowledge management, and I briefly described each as a field of study, and assessed their defining characteristics. I concluded that the three concepts and fields of study are closely related, that they deal with similar and overlapping concerns, but have distinct areas of

133

focus. Organizational learning is the most general concept of the three; it is primarily focused on defining what organizational learning is, on different types of organizational learning, on learning processes present in organizations, and on the question of when organizational learning takes place. The learning organization concept and its related field of study, considered by Argyris and Schon (1996) to be a branch of organizational learning literature, focuses primarily on the organizational learning environment, assessing the strengths and weaknesses of this environment in a given company or organization, and on how to create conditions in which organizational learning will occur. Knowledge management meanwhile is most directly concerned with questions pertaining to business applications and outcomes of knowledge such as where and how tacit knowledge is embedded in company routines; how to operationalize that knowledge; how to transfer tacit knowledge into more explicit forms; how to use tacit knowledge to generate new forms of explicit knowledge; and, in general, how to leverage tacit knowledge as a firm-specific resource. Rather than trying to establish whether or not organizational learning is taking place in the company, the research design for this study focused on revealing the learning processes— whether complete or partial. Therefore, like in the literature, organizational learning and learning organization concepts do overlap considerably in the results of this study. As discussed in the relevant section, the survey results suggest the company is weaker when it comes to learning processes and practices compared to the learning environment; hence the survey results can help the company assess its own learning processes and practices, and to develop measures to address insufficiencies. For example, the company can insure that certain key competencies and knowledge are rendered in an explicit and retrievable form so that they do not leak from the company when individuals who possess these key skills and knowledge leave the enterprise.

134

Interpreting the results in this way supports the literature on the learning organization, particularly the literature focused on rendering the concept empirically measurable and applicable (An & Reigeluth, 2005; Bui & Baruch, 2010; Garvin et al., 2008; Heorhiadi et al., 2014; Jagasia et al., 2015; Kirwan, 2013; Lazar & Robu, 2015; Shipton et al., 2013; Yang et al., 2004). Furthermore, in the field of organizational learning, the four criteria for group learning proposed by Wilson et al. (2007)—that the learning take place on a group level, that there be fundamental processes present such as retrieval, that the learning produce an outcome, and that learning take place over a time interval—can be considered a model for the company to strive for. Argyris and Schon (1996) argue that one of the weaknesses of the learning organization literature is that it assumes all learning in an organization is good; in other words, that there is little critical questioning of the content of the learning. Consistent with this critical view, the results of this study suggest that not all learning taking place in the company is necessarily positive. For example, in Group Interview 3, one of the employees uses the phrase “free style” to convey that he and his colleagues often feel compelled to work at breakneck speed and to improvise in order to deliver e-commerce sites or features of sites within the promised deadlines, hence skipping standard procedures. This increases the risk of delivering sites with technical problems that will have to be dealt with after going live. The whole group participating in the interview seemed to share this perception; this suggests that employees—at least those involved in the e-store activity—learn to work this way, whether or not this is intentional on the part of the company’s directors. Such learning may be happening more by neglect than by design. If this type of learning, which runs the risk of being counterproductive in the long run, becomes solidly

135

embedded in the company’s culture, it may form what Christensen (1997/2011) refers to as a “disability” rather than an ability. The knowledge management literature is also consistent with a critical approach to the type of learning taking place. Managing knowledge implies the company should be conscious, intentional, and selective in how it addresses the range of skills, practices, and knowledge present in the company; this implies that existing productive skills and learning processes should be identified, encouraged, and leveraged, while unproductive ones should be eliminated or corrected. The semi-structured interview results described in the Causal mapping analysis and Knowledge-in-practice framework analysis subsections are consistent with this approach because they reveal a number of tacit skills present in the company which the company’s leaders may not have been aware of previously, and because jobs and activities are mapped out according to a framework which facilitates developing effective knowledge management strategies and tools. In conclusion, the three interrelated concepts and fields of study can be used to analyze the same problems and questions. The different perspectives and implications for action allow for a more complete understanding of the relevant knowledge and learning issues, and suggest a more complete range of possible actions to address them. Company initiatives inspired by the results of this study, such as actions taken to reduce knowledge leaks when individuals leave the company, may be simultaneously understood as knowledge management initiatives, organizational learning initiatives, and/or steps taken by the company towards the goal of becoming a learning organization.

136

Research question b: What are different types of organizational knowledge and how are they related? The main types of organizational knowledge dealt with in this study are tacit and explicit organizational knowledge. As stated earlier, Ambrosini and Bowman (2001) prefer the phrase, “tacit skills” to “tacit knowledge”, because they argue that tacit skills are anchored in practice, hence not abstract like the word “knowledge” implies. Organizational tacit skills are thus things that individuals and/or groups in the organization know how to do, but which have not been rendered in any explicit form—such skills are retrievable only in so far as people remember them. Explicit knowledge, a more abstract term, refers to bodies of knowledge which the company possesses, thanks to explicit training and education of its leaders and employees, and/or it refers to know-how which is stored, retrievable, and thus can be easily shared. The two categories of tacit and explicit knowledge, however, are very broad and can be further broken down into other types of knowledge. As noted in the Knowledge management subsection in Chapter 2, Spender (1993) identifies four types of organizational knowledge: conscious, automatic, scientific, and communal—the first two are individual, and second two are collective; he considers all of them to be tacit except scientific knowledge. The distinction between conscious and automatic individual tacit knowledge appears to be consistent with Ambrosini and Bowman’s degrees of tacitness (2001, p. 816), referred to in the Chapter 3 subsection Semi-structured interview protocol; conscious individual tacit knowledge corresponds to the level of tacit skills that can be “imperfectly articulated”, while automatic knowledge corresponds to “deeply ingrained tacit skills”. As explained in Chapter 3, the causal mapping method in this study aims primarily to reveal conscious individual tacit knowledge (skills that can be “imperfectly articulated”), and the

137

results described in the subsection Causal mapping analysis are consistent with this aim. Skills such as anticipating trends, communication skills, and people management skills are examples. Communal tacit knowledge, also revealed in this study, would include the organizational level skills such as managing work flows or communicating across departments—the results of the research suggest these skills are inconsistent, partial, or, in some cases, insufficient. Meanwhile, scientific knowledge (hence, explicit knowledge) in the company includes the crucial technical skills and knowledge which allow the organization to design, create, and maintain e-commerce sites, as well as to analyze traffic, and to conduct financial analyses. While not the object of this study, the scientific knowledge is, of course, necessary for the company to function. Because there is an emphasis on action and routines in the understanding of tacit knowledge as skills anchored in practice, an underlying question in this section is whether or not practice itself can be considered a form of organizational knowledge. As already discussed in the Chapter 2 subsection Knowledge management, Cook and Brown (1999) propose that the epistemology of knowing, which they consider to be a dynamic and relational concept close to practice, be given as much importance in the literature as the epistemology of knowledge, which they consider a more static concept. Furthermore, Spender (2014) proposes to define the firm as “an integrated community of skillful practices” (p. 8). The results of this study, which reveal skills embedded in their specific contexts, are consistent with an understanding of practice as a dynamic type of knowledge. Research question c: How can firm-specific tacit knowledge of employees in an organization be mapped to help identify crucial organizational skills? This research question can be answered by analyzing the results generated by the two analytical tools which were used to design the semi-structured interviews, and to organize the

138

interview data—the causal mapping method proposed by Ambrosini and Bowman (2001), and the knowledge-in-practice framework developed by McIver et al. (2013). As noted in Chapter 3, the causal mapping method is designed to operationalize tacit knowledge (skills). Applying this mapping technique means identifying important constructs and factors of organizational success, exploring their definitions in the organization, and generating examples. Tacit skills are revealed—albeit imperfectly and partially—when employees provide examples, narratives, or metaphors which illustrate causes (or enabling factors) of organizational success constructs. As described in the subsection Causal mapping analysis, the results yielded numerous tacit skills which are highly firm-specific. Because these skills are related to important organizational constructs such as innovation, and good relations with the clients, the skills are likely to be crucial to the organization and/or linked to other skills which are crucial to the organization. The knowledge-in-practice framework meanwhile characterizes specific positions, and/or practices in the company according to their respective degrees of tacitness and learnability. The results described in the subsection Knowledge-in-practice framework analysis allow for positions and practices to be analyzed in relation to each other; furthermore, Figure 17 includes arrows to indicate when positions have characteristics of more than one KIP type which helps to make it clear what types of crucial organizational skills are associated with specific positions, Figure 18 then helps identify a wide range of company skills (distinct from any specific positions) and indicates which KIP types they best correspond to. Overall, the results from the semi-structured interviews help to map and assess the skills present in the company with a view to better understanding them, and to more effectively leveraging them. While the mapping process is partial rather than exhaustive, it can be completed by more systematic efforts on the part of the enterprise.

139

Research question d: What types of practices in the company contribute to the creation of new knowledge, and how might the company leverage those practices to develop new explicit knowledge which can potentially be leveraged as a sustainable competitive advantage? One way to address this research question is to analyze the results of the semi-structured interviews using Nonaka’s four modes of knowledge creation, described in Chapter 2: socialization, combination, externalization, and internalization (1994). These modes of knowledge creation are conceived as a continuous spiral; Nonaka (1994) states that, “organizations continuously create new knowledge by reconstructing existing perspectives, frameworks, or premises on a day-to-day basis” (p. 19). According to this theory, interactions, which may be non-verbal (i.e. observation and imitation) facilitate the sharing of tacit knowledge among individuals—this is the socialization mode. The combination mode transpires when meetings and conversations in the enterprise afford the opportunity to employees to combine or reconfigure existing bodies of explicit knowledge into new knowledge. According to Nonaka, externalization then refers to the process by which tacit knowledge is converted into explicit knowledge; while internalization, conversely, refers to the process of converting explicit knowledge into tacit knowledge, which is to say embedding it into practice. The results of this study point to a number of existing practices, emerging practices, and potential practices which contribute (or have the potential to contribute in the future) to a virtuous knowledge creation cycle. Several positions in the company, which were analyzed in the section Results from Semi-structured Interviews, require on-the-job learning and practice in order to hone skills. Exchanges with co-workers and feedback from managers are key to this type learning and interaction. Hence, employees do have opportunities to learn tacit skills from

140

each other in a process of socialization. Furthermore, team meetings and/or departmental meetings provide the context for combining knowledge into new knowledge in a process of combination. Therefore, two modes of knowledge creation appear to be present in the company. It is unclear, however, the extent to which externalization and internalization processes are taking place; this is most likely linked to the apparent weaknesses in concrete learning processes and practices identified in the results of the Learning Organization Survey (Garvin et al., 2008). Nevertheless, the potential is there to identify examples of productive tacit knowhow and to spread this know-how more evenly across the company; hence the externalization mode can be reinforced (if existing at all) or introduced (if currently inexistent). For example, managerial practices identified as effective at developing employee autonomy can be taught to other managers. Strategies which allow certain employees to face situations characterized by a high degree of uncertainty can be shared among colleagues and/or rendered in an explicit form such as a rubric or written guide. Metaphors, such as “opening the hood of the car to see how the engine works”, can be used to communicate complex concepts like thinking holistically to all employees, or to targeted groups of employees. Internalization will then transpire if the company's leaders design and implement effective strategies for making sure the desired practices do in fact spread and become more systematically embedded in the company's way of doing things. As noted in the literature review, Cook and Brown (1999) have a different view on types of knowledge and how they interact to create new knowledge; these authors emphasize the generative and dynamic dimension of knowledge creation rather than the notion of knowledge transfer. Practice is at the center of their epistemology of knowing because it continuously informs other forms of knowledge (Cook & Brown, 1999). According to this view, if the

141

company develops explicit knowledge—for example, a manual detailing how to set up a payment system—this is not a matter of transferring an individual's tacit knowledge gleaned from experience, to an explicit, static form; but it is rather an example of knowledge creation. The individual uses his tacit skills and accumulated experience, possibly in collaboration with other employees, to develop something new—a set of instructions. While it may seem like a question of nuances, the advantage of this view is that it places practice at the center of the theory and encourages a constantly evolving understanding of knowledge creation. This is important for a company which must remain agile to survive in a constantly changing sector. Research question e: How can learning be assessed in an organization with a view to identifying strengths and weaknesses? This research question is addressed primarily with the Learning Organization Survey (Garvin et al., 2008) described in the relevant Chapter 3 subsection, the results of which are analyzed in the section Survey Results. The survey is intended to measure organizational learning occurring in any organizational unit; the authors provide benchmark scores which allow for comparison to the combined company results; the company results can also be compared internally across organizational subunits. The results of this survey help to assess the strengths and weaknesses of learning in the company; the results of the semi-structured interviews provide context to better make sense of the survey results and confirm the survey’s findings. Consistent with the intended use of the Learning Organization Survey (Garvin et al., 2008), and the relevant learning organization literature, the survey results clearly suggest strengths and weaknesses, which are described in detail in the section Survey Results. For example, the results provide evidence that the company is weak in the second building block, Concrete Learning Processes and Practices, particularly for the two items, Education and

142

Training; and Information Transfer. The fact that the findings are similar when comparing the company's medians to the benchmark medians, when comparing the company’s means to the benchmark medians, and when analyzing the standard deviations for the survey items, suggests the findings are reliable. The results of the semi-structured interviews not only provide context for the survey results but also tend to confirm the findings. For example, in the Causal mapping analysis subsection, where the results of the causal mapping method are analyzed, specific situations related by employees are described in which training appears to be lacking and/or in which the transfer of information appears to be deficient. Other examples which confirm and/or provide context for the survey findings can be found in the interview data, particularly regarding the two organizational constructs good relations with clients; and client and employee development. These examples provide insights about why the employees perceive the company to be weak in concrete learning processes. Furthermore, the examples point to specific areas and situations which reveal potential weaknesses and hence can serve as starting points for envisaging initiatives such as training programs. Moreover, following the results analyzed in the Knowledge-in-practice framework analysis subsection, using the knowledge-in-practice framework can help ensure that knowledge management initiatives are well matched to specific positions and/or practices in the firm. Summary of Major Findings The results from the Learning Organization Survey (Garvin et al., 2008), previously detailed in the Survey Results section, show an emergent learning organization culture in the company, but one that has much room for reinforcement. The external benchmark comparisons suggest some company strengths in the first building block, Supportive Learning Environment,

143

particularly the item, Appreciation of Differences. The learning environment therefore appears supportive of differences of opinion and alternative ways of getting things done. The company is also reasonably strong in Psychological Safety, suggesting that employees feel comfortable speaking their minds, and that they feel they can make mistakes without necessarily being punished. On the other hand, the low comparative results for the item, Openness to New Ideas, may suggest that despite a learning environment which supports different ways of doing things, and supports differences of opinion, there is a lack of effective listening when it comes to the sharing of new ideas. Meanwhile, the benchmark comparisons indicate weaknesses in the second building block, Concrete Learning Processes and Practices, across nearly all the items, and particularly for the two items, Education and Training; and Information Transfer. Broadly speaking, the results suggest the company does not do enough beyond cultivating an environment favorable to learning; there is an insufficient framework in place for a more deliberate and systematic approach. In particular, there is insufficient training for new and existing employees, and there are too few organized activities and processes to share information internally and externally. The results of the internal comparisons of standard deviations per item, and of the standard deviations between departments and levels of hierarchy are largely convergent with the benchmark comparisons, especially regarding the weak results in the second building block. The same items that have low comparative scores in the second building block, Education and Training, and Information Transfer, also have the lowest absolute scores, and show the highest standard deviations, which suggests a lack of consensus among respondents on these items. The results from the semi-structured interviews, which are analyzed according to the causal mapping methodology in the relevant subsection, and analyzed according to the

144

knowledge-in-practice framework in the relevant subsection, reveal tacit skills and furthermore provide context for interpreting the survey results. The causal mapping analysis reveals a number of important tacit skills associated with key organizational constructs and success factors. Skills associated with innovation, for example, include the ability to anticipate trends, and to think holistically and creatively, as well as the ability to create positions and departments in order to exploit new activities. Skills associated with commercial success include the ability to convince clients, and to learn and improve offers thanks to accumulated experience. Skills associated with maintaining good relations with clients include individual, group and organizational communication skills, as well as team work skills. Tacit skills associated with employee and client development include people management skills such as delegating and asking the right questions, and also include information transfer skills such as knowing what information to share with the client, and how and when to share it. Many examples which were elicited thanks to the causal mapping methodology provide insights useful in the interpretation of the results of the Learning Organization Survey (Garvin et al., 2008). For instance, examples illustrate how a lack of systematic company training and transfer of information among employees conflict with key organizational constructs such as maintaining good relations with clients. Likewise, some employees expressed frustration with their lack of perceived autonomy and lack of freedom to take initiative which suggests that a more systematic approach is needed for training managers in order to realize the key organizational construct of employee development. Furthermore, employees appear to be left largely on their own when it comes to deciding what information to share with the client, what client development actually means, and how to achieve it. These examples therefore provide

145

evidence which converges with the weaknesses identified in the Survey Results section. The examples also allow for a richer and more complete understanding of these weaknesses. The analysis of results using the knowledge-in-practice framework demonstrates it is not easy to reveal the tacit dimensions of several key positions in the company. With the exception of the co-founder/head of photography, the positions appear, at first view, to be dominated by low tacit skills which are difficult to learn because of the importance placed on mastering complex, specialized information. This masks the reality, however, that there is a significant amount of tacit knowledge and skill inherent in these same positions because of the dynamic, interactive, and evolving nature of the activity. Most employees and managers must learn on the job constantly, hone their skills thanks to feedback and exchanges with colleagues, and are frequently expected to take the initiative. Important company developments which came to light during the semi-structured interviews were the recent re-orientation of the company’s commercial strategy away from a full-service offer to a more tailored services offer, and a re-orientation of the company’s innovation strategy to focus more on technological/product innovation (i.e. developing an ecommerce platform for the apparel sector). The re-orientation of the commercial strategy appears to require building a rational, analytical framework on top of the existing base of tacit knowledge. It is thus a good example of what Christensen (1997/2011) describes when he states that companies over time move towards more visible and explicit processes (p. 195). Furthermore, the product innovation orientation requires the company to render in an explicit, stored, and accessible format, tacit knowledge accumulated through experience.

146

These major findings from the survey and semi-structured interviews are discussed in more depth in terms of their relevance to theory and practice in Chapter 5: Discussion, Recommendations, and Conclusions.

147

Chapter 5: Discussion, Recommendations, and Conclusions Synopsis Scope of the study. This study focused on two areas of needed research: e-commerce, and the development and leveraging of tacit knowledge for competitive advantage. The two main types of e-commerce discussed in this research are business-to-consumer, and business-tobusiness—with the latter representing the far greater share of overall e-commerce. The two terms cover a wide diversity of digitally enabled commercial transactions such as consumer-toconsumer e-commerce, online market places, local e-commerce, and social e-commerce. Ecommerce can be distinguished from brick-and-mortar commerce by the universal standards involved in Internet technology, the potential for global reach and ubiquitous presence available to companies, the richness of the medium and the density of the information available online, the potential for customizing the messages, and by its social dimension, which has developed over the past ten years thanks to the Web 2.0 (Laudon & Traver, 2017). The history of e-commerce since the mid 1990’s has been a complex and dynamic mix of technological breakthroughs, infrastructure development, evolving standards and regulations, investments and innovations by entrepreneurs and venture capitalists, and, finally, of changing consumer behavior and expectations. Parallels can be made to typical stages of development of earlier revolutionary technologies such as telegraphy, automobiles, and radio (Varian, 2001). Moreover, e-commerce can be viewed in the context of the Age of Information and Telecommunications (Perez, 2009a, 2009b). Recent and current developments in e-commerce include the use of mobile phones to make transactions, and the importance of user-generated content and social networking sites. Meanwhile, cryptocurrencies, artificial intelligence, and virtual reality represent the most recent trends and point towards the future of the sector. Despite

148

the obvious importance of technology in e-commerce, Orlikowsky and Iacono’s (2001) definition of the digital economy as “an ongoing social product, shaped and produced by humans and organizations” (p. 357) should be born in mind along with Christensen’s (1997/2011) argument that innovation is more about processes and values than resources. There is thus a need for studies which assess workplace practices and routines in the e-commerce sector, and which attempt to draw links to firm-specific advantages and long-term, sustainable competitive advantage. Tacit knowledge is also in need of further study. By definition, it is emergent knowledge that is difficult to describe and conduct research on because it is embedded in individuals and deeply rooted in specific contexts. Ambrosini and Bowman (2001) prefer the term “tacit skills” to “tacit knowledge” because the construct deals with know-how and practice more than abstract knowledge. Tacit knowledge has the potential to emerge as a source of competitive advantage precisely because it is context specific, and therefore, difficult to imitate, transfer or substitute. According to Polyani (1966), tacit knowledge is more fundamental than explicit knowledge; it can be considered a raw base of knowledge from which explicit knowledge stems. Ambrosini and Bowman (2001) argue for the existence of degrees of tacitness, and assert that, while deeply ingrained tacit skills are inaccessible to the researcher, less deeply ingrained tacit skills can be articulated during interviews. In this way, a crucial resource of the firm can be operationalized, and its potential relationship to the firm’s long-term competitive advantage appraised. The study company was chosen for this research because it is an example of a young, innovative technology company whose capabilities are migrating from resources such as key people towards “visible, conscious processes and values” (Christensen, 1999/2011, p. 195). Founded in 2008 by three young professionals, the enterprise started as a full-service business-

149

to-business e-commerce company for apparel brands representing the mid-level, premium and entry luxury markets. The company has grown rapidly; at the time of this study, it employed 115 staff members whose training and expertise cover both e-commerce technology and marketing. Other than the members of the executive team, and the top managers, the employees are primarily technical employees and/or lower-level managers who have less than five years of professional experience. Recently the company has been working to diversify its business model and to debundle its services so that the full-service e-commerce offer is proposed as one option among others. In addition, the company is developing its own e-commerce platform which it intends to be more adapted to the apparel sector than existing e-commerce platforms; their plan is to license the platform to clients. Theoretical background. In recent years, knowledge management has emerged as an important strategy to acquire and maintain competitive advantage in organizations engaged in dynamic industries such as e-commerce. In turn, knowledge management combines work in areas such as the learning organization, and organizational learning. The learning organization concept may be considered a strategic goal for developing a company environment favorable to learning, and for developing concrete learning practices and processes (Aggestam, 2006; Garvin et al., 2008; King, 2001; Rowden, 2001). Senge (1990/2006) argues that five disciplines are key in building learning organizations: systems thinking, personal mastery, mental models, shared vision, and team learning. Organizational learning, meanwhile, is a discipline concerned with how organizations change as they acquire experience; it developed out of the disciplines of behavioral psychology and organizational behavior, and it deals with creation, retention and transfer of knowledge within an organization, as well as with learning processes and outcomes (Argote, 2011; Argote & Miron-Spektor, 2009; Argyris & Schon, 1996; Dodgson, 1993).

150

According to Prusak (2001), knowledge management emphasizes concrete, situationspecific value of knowledge rather than theoretical, abstract knowledge. The works of Cook and Brown (1999), Nonaka (1994), and Spender (1993) meanwhile describes how tacit, explicit, individual, and collective knowledge, not to mention practice, can interact in dynamic knowledge creation cycles which contribute to competitive advantage. The fields of organizational learning and knowledge management overlap considerably; both are multidisciplinary to a significant extent. Findings from within the different approaches may be cumulated to advance our overall understanding of the related phenomena of knowledge and learning the workplace (Argote, 2005). The review of the literature produced three bodies of work used to anchor the theory underlying this study and suggest a workable methodology. The Learning Organization Survey developed by Garvin et al. (2008) is a survey tool to assess learning practices in the company. As a group-level measurement instrument, it is intended to measure learning which occurs in “an organizational unit of any size that has meaningful shared or overlapping work activities” (Garvin et al., 2008, p. 110). Also, the work of Ambrosini and Bowman (2001) proposes a causal mapping method for operationalizing the concept of tacit knowledge. Causal mapping attempts to reveal tacit skills of participants during semi-structured interviews by asking the right questions and eliciting narratives about daily practices in the workplace; furthermore, causal relationships may be appraised between key organizational constructs and the tacit skills elicited. Finally, the work of McIver et al. (2013) provides an additional framework for the semistructured interviews, and for the analysis of the interview data. Their knowledge-in-practice framework (McIver et al., 2013) is meant to identify degrees of tacitness and learnability of work

151

activities in organizations in order to help researchers and practitioners design more effective knowledge management planning and application strategies. Methodology. Tacit knowledge is by definition difficult to study. Therefore, this study used a mixed model of qualitative and survey methods to address the research questions. The Learning Organization Survey (Garvin et al., 2008), a cross-section descriptive survey, was administered to each of the 41 participating employees of the company before each semistructured interview (conducted individually for the top managers, and in groups for the other employees). The survey is designed assess the extent to which the enterprise functions as a learning organization; and to evaluate the relationships among the factors which affect organizational learning (Garvin et al., 2008). The survey is built on the theory that there are three major factors, referred to as “building blocks”, which must be present for organizational learning to occur: a supportive learning environment; concrete learning practices; and leadership that reinforces learning. To complete the survey, employees rate how accurately the statements—grouped under each of the three building blocks—match their perceptions of the organization they work for, and its learning culture. The quantitative results can be compared to benchmark scores provided by the survey’s authors, as well as compared internally across departments and levels of hierarchy. Semi-structured, open-ended interviews were conducted with both individuals and small groups immediately after each participant completed the survey. Ambrosini and Bowman’s (2001) causal mapping methodology was applied during the semi-structured interviews with a view to asking questions to elicit tacit skills which can be articulated, and to get participants to tell stories and/or make use of metaphors to talk indirectly about tacit skills which can only be imperfectly or indirectly articulated. Having first identified constructs related to organizational

152

success during pre-interviews and/or early in the interview process, I started the causal mapping process by asking questions about how factors related to the constructs occurred in the enterprise, and what enabled them to happen; also I attempted to elicit specific examples and relevant narratives whenever possible. The knowledge-in-practice framework, developed by McIver et al. (2013), was also integrated into the semi-structured interview protocol to assess the underlying knowledge structure of the work activities described by the participants along the two dimensions of tacitness and learnability. I asked questions about what constituted high performance for specific jobs, how this was achieved, and the knowledge required to achieve it. This allowed me to not only map causal relationships between organizational success constructs and enabling factors (such as tacit skills) but also to map the work activities described according to four practice types proposed by McIver et al.: enacted information (high learnability, low tacitness), accumulated information (low learnability, low tactiness), apprenticed know-how (high learnability, high tacitness), and talent and intuitive know-how (low learnability, high tacitness). The quantitative and qualitative results were integrated in order to effectively address the research questions in this study. The semi-structured interview data revealed tacit skills and causal relations between constructs and enabling factors, and they provided insights about learning practices in the enterprise which allowed for a richer understanding of the survey data. Strengths and weaknesses revealed by the survey scores in the three building blocks, and/or in the subcomponents of each building block, were contextualized and confirmed by analyzing recurrent themes in the interview data. The specific problems faced by participants, and how they described facing those problems provided a level of detail based on lived experience inaccessible to quantitative survey data. The knowledge-in-practice framework (McIver et al.,

153

2013) allowed for an augmented understanding of the relationship between tacit skills in the company, on the one hand, and specific positions and work activities, on the other; this helped contextualize the internal and external comparisons of the survey. Results and answers to research questions. The principal research questions and results of the study are summarized here. Research question a: How are the concepts of organizational learning, knowledge management, and the learning organization related? The three fields of social sciences literature which contributed to the theoretical foundation of this study are closely related. They deal with similar and overlapping concerns; however, they each have distinct origins and areas of focus. The concept of organizational learning and its corresponding literature is the broadest of the three because it deals with definitions and criteria for group and organizational learning, as well as describing and analyzing organizational learning processes. While stemming from the field of organizational learning, the learning organization concept is closer to a strategic model; the emphasis in much of the literature is on empirically measurable applications for organizational development. Knowledge management is similarly concerned with business applications but takes a broader perspective which includes assessment of different types of knowledge in the enterprise, and how these types of knowledge interact to potentially generate new knowledge. The results of the study can be interpreted in ways that are coherent with all three fields. Compared to any one single approach, the different concepts and corresponding literature complement each other to provide a more complete understanding of the organizational knowledge and learning issues relevant to this study; moreover, they suggest a wider array of possible actions for the company.

154

Research question b: What are different types of organizational knowledge and how are they related? Organizational knowledge is both tacit and explicit. Explicit organizational knowledge refers to the abstract bodies of knowledge which the company possesses thanks to the training and education of its members, and to the know-how which the company has stored, can easily retrieve, and can disseminate throughout the organization. This latter type of explicit organizational knowledge is more specific to the firm than the general abstract knowledge. Tacit organizational knowledge, which may be automatic, conscious, and communal, refers to knowhow that has not been rendered in any explicit form but is embedded in practices and routines, and is thus very rooted in specific contexts. Furthermore, practice may be considered a dynamic form of knowledge that links the other types of knowledge, and which contributes to knowledge generation cycles. Individual conscious tacit skills, as well as organizational tacit skills specific to the enterprise, were revealed in this study. In assessing this study findings, I explore the relationship between the existing explicit and tacit organizational knowledge; I also explore the need to develop more explicit forms of know-how from the existing tacit knowledge. Research question c: How can firm-specific tacit knowledge of employees in an organization be mapped to help identify crucial organizational skills? The causal mapping method (Ambroini & Bowman, 2001), and the knowledge-inpractice framework (McIver et al., 2013) were both used to reveal and map firm-specific tacit skills in order to identify crucial organizational skills possessed by the company; these skills were associated with organizational success constructs. For example, the ability to anticipate trends, and to think holistically and creatively, as well as the ability to create positions and departments to exploit new activities were all skills identified in this study associated with

155

innovation. The ability to convince clients, and to learn and improve offers thanks to accumulated experience were skills identified in this study were associated with commercial success. Skills associated with maintaining good relations with clients included individual, group and organizational communication skills, as well as team work skills. Finally, tacit skills associated with employee and client development included people management skills, such as delegating and asking the right questions, and also included information transfer skills such as knowing what information to share with the client, as well as how and when to share it. The knowledge-in-practice framework (McIver et al., 2013), while revealing that many positions in the company appear to be dominated by low tacit skills which are difficult to learn, also revealed a significant amount of tacit knowledge and skill inherent in these same positions because of the dynamic, evolving nature of the activity. For example, employees and managers must be able to learn on the job constantly, hone their skills thanks to feedback and exchanges with colleagues, engage in holistic business analysis, and take the initiative. While the mapping process is partial rather than exhaustive, it can be completed by more systematic efforts on the part of the enterprise. Research question d: What types of practices in the company contribute to the creation of new knowledge, and how might the company leverage those practices to develop new explicit knowledge which can potentially be leveraged as a sustainable competitive advantage? The findings suggest that employees do have opportunities to learn tacit skills from each other in processes of socialization, many of which are informal. Team meetings and departmental meetings also provide the context for combining knowledge into new knowledge in a process of combination (Nonaka, 1994). In addition, the results suggest that some brand

156

managers are effective at developing the autonomy of their e-store teams by asking them good questions, and by delegating decision making power. Further practices were revealed related to client development such as providing expert advice, and sharing information. Some employees have developed the ability to face high levels of uncertainty and to solve complex problems; one such employee used the metaphor of “opening up the hood of a car to understand how the engine works” as a way to express the curiosity, persistence and holistic thinking required to face complex problems. The enterprise must be more systematic in identifying best practices, and in disseminating them throughout the organization. This can be done by reinforcing existing socialization processes with more formal processes to effectively capture, store, and disseminate best practices. Furthermore, disseminating best practices is likely to require training, mentoring, and the use of metaphors and/or images to communicate the company’s expectations and values. A complete knowledge creation cycle, according to Nonaka (1994), will develop only when the company implements externalization and internalization processes to complement the existing socialization and combination processes. Research question e: How can learning be assessed in an organization with a view to identifying strengths and weaknesses? The study primarily makes use of the Learning Organization Survey (Garvin et al., 2008) to assess learning in the study company, and to identify strengths and weaknesses. The results of the survey were compared externally to the benchmark scores provided by the survey’s authors, and they were compared internally across departments and levels of hierarchy. The results of the semi-structured interviews provide additional data as well as context for analyzing the strengths and weaknesses in the company’s learning processes revealed by the survey scores.

157

The external benchmark comparison reveals company strengths in the first building block, Supportive Learning Environment, particularly in the item, Appreciation of Differences. The learning environment thus appears supportive of differences of opinion and alternative ways of getting things done. The company also appears strong in Psychological Safety, suggesting that employees feel comfortable speaking their minds, and that they feel they can make mistakes without necessarily being punished. The benchmark comparison, meanwhile, demonstrates weaknesses in the second building block, Concreate Learning Processes and Practices, across nearly all the items and, particularly for the two items, Education and Training; and Information Transfer. The results of the internal comparison are largely convergent with the benchmark comparisons, especially regarding the weak results in the second building block. Consistent differences can be noted, however, between the scores from the human resources department, and the e-commerce department—the former being higher than the latter. The results of the semi-structured interviews tend to confirm the survey findings. Specific situations related by employees are described in which training appears to be lacking, and/or in which the transfer of information appears to be deficient. These examples provide insights about why the employees perceive the company to be weak in concrete learning processes. Furthermore, the examples point to specific areas and situations which reveal weaknesses, and hence can serve as starting points to develop initiatives such as training programs. Finally, the results analyzed using the knowledge-in-practice framework can help ensure that knowledge management initiatives are well matched to specific positions and/or practices in the firm. With the above synopsis as background, the results in terms of theory, practice, and future research will now be discussed.

158

Discussion This section addresses the implications of the study findings in light of current theory, as well as in light of issues pertaining to measurement, and research; lastly, it discusses implications for improving practice in the company, and in the sector of e-commerce. The section is divided into four subsections. The Theory subsection explores links between the literature on the life cycle stages of company development, on the one hand, and the three fields in the management and social sciences literature already discussed in this study (organizational learning, learning organization, and knowledge management), on the other. This subsection also explores themes such as the company’s readiness for change, knowledge generation and dissemination within the enterprise, the role of tacit information in the development of organizational culture, the role of tacit information in the development of successful teamwork, and how teamwork contributes to organizational learning. The Measurement subsection then evaluates the utility of the causal mapping method in mixed-methods research, and its effectiveness in capturing phenomena which are difficult to quantify such as tacit learning, organizational culture, and teamwork. Following this, the Research subsection outlines opportunities for future research on different types of learning germane to the quest for competitive advantage, and it explores possible research directions which might advance our understanding of e-commerce. Finally, the Practice subsection looks at ways in which the study findings might be used by the company to improve internal processes and effectiveness, as well as ways to improve the functioning of e-commerce companies, in general. Theory. This theoretical discussion starts by reviewing the literature on organizational life cycles and stages of development; it explores links between this field of research, on the one hand, and the social sciences literature on organizational learning, the learning organization, and

159

knowledge management, on the other. The applicability of life cycle theory to the study company, and to the e-commerce sector is evaluated. Also, the study results are assessed in light of relevant theory on organizational life cycles. The next part of this discussion appraises the lack of internal homogeneity in the survey results across departments and levels of hierarchy in the context of a strategic learning organization model which emphasizes readiness for change. This is followed by an assessment of how knowledge gets generated and disseminated within the organization—an assessment which is facilitated by applying cyclical models of knowledge creation and learning processes to the study company, as well by considering effective strategies for facilitating the adoption of useful innovations on an institutional level. After this, there is an analysis of the role of tacit knowledge and skills in the development of organizational culture; here the study findings are once again appraised in the context of relevant theory. The last part of this discussion explores the role of tacit information in the development of successful teamwork, how team work contributes to organizational learning, and what the study results reveal about team work in the study company. Finally, the conclusions of the theoretical discussion are summed up. Life cycle stages of development and the link to organizational learning in e-commerce companies. The concept of organizational life cycles has developed since the 1960s and has gained widespread acceptance in scholarly literature (Quinn & Cameron, 1983; Smits & Bowden, 2015). Greiner (1972), for example, proposes a dynamic concept of developmental phases which depend on five key dimensions: the age of the organization, the size of the organization, stages of evolution (periods of stable growth), stages of revolution (periods of substantial turbulence), and the growth rate in the industry (p. 38). According to this theory, management practices that work in one phase of development lose their effectiveness as the

160

organization grows, which eventually leads to a leadership crisis. When changes in leadership and management practices are achieved successfully, the company has transitioned into the next phase of development (Greiner, 1972). In a later article which synthesizes nine different life cycle models (including Greiner’s), Quinn and Cameron (1983) emphasize the criteria of organizational effectiveness as drivers of change, and they suggest these criteria depend largely on the power of constituencies. For example, while a startup organization in its initial stages of development is likely to be dominated by its founder and a small group of trusted people, later in the company’s development, investors or state actors may play a greater role and impose completely different criteria for organizational effectiveness—this creates a leadership crisis. Summing up the literature on this topic, Smits and Bowden (2015) state that “over time, startup organizations face similar crises and challenges, and grow and mature in predictable ways as they survive and prosper” (p. 5).

161

Figure 19. Organizational stages of development (Daft, 2010, adapted from Quinn &Cameron, 1983 and Greiner, 1972) Figure 19 shows Daft’s (2010) life cycle model which adapts the four-phase model developed by Quinn & Cameron (1983), and integrates the dynamic concept of evolution and revolution developed by Greiner (1972). The four stages in Daft’s model are: entrepreneurial, collectivity, formalization, and elaboration. The entrepreneurial stage is characterized by creativity and survival; production and marketing; personal supervision by the founders; long working hours; and little formality or coordination. The collectivity stage is characterized by a strong sense of mission and commitment; staff and workers feel that they are part of a collective. While most communication and control remain informal, some formal systems start to appear

162

such as departments, job assignments, and a division of labor. In this stage, the role of the founder/owner is less significant—some of the active management role may be delegated to professional managers. The formalization stage emphasizes efficiency and control; it is characterized by the formalization of rules and structure, conservatism and institutional procedures. Finally, the elaboration stage occurs when the organization has to face a changing business environment; it is characterized by the elaboration of structure in order to achieve continued growth—this may take the form of diversification, the creation of cross-functional teams, and/or the formation multiple company divisions. There is very little research available on stages of development specifically for ecommerce companies. As discussed in Chapter 2 in the section Introduction to the E-commerce Industry, there are numerous characteristics of e-commerce which differentiate it from traditional commerce. Examples include its ubiquity, global reach, universal standards, richness, interactivity, information density, customization, and social technology. At the same time, ecommerce can be considered part of the larger phenomenon referred to as the Digital Economy (Varian, 2001) or the Age of Information (Perez, 2009), which scholars have demonstrated follows cycles similar to prior technological revolutions. Furthermore, as stated in the relevant Chapter 2 section, Laudon and Traver (2016) suggest that e-commerce itself follows business cycles including experimentation, capitalization, hyper-competition, and consolidation. Microsoft and Facebook are both provided as examples by Daft (2010) of companies who have passed through the stages of development. While neither company exactly fits the definition of an e-commerce company, and they both dwarf the size of the study company, their activities are highly relevant to the sector, and they provide examples of companies which grew rapidly in a complex and evolving business environment. Therefore, despite some of the unique and

163

unprecedented aspects of e-commerce, there is no compelling reason to assume that life cycle models would not apply to the sector, and to the study company as they would to companies in other sectors. Although a young entity, the study company fits many of the characteristics of the collectivity stage (Daft, 2010). Two of the original founders of the company have removed themselves from active management roles, and professional managers have been hired to deal with operations, e-commerce, human resources, and finance. Interviews reveal a sense of collectivity and intense commitment on the part of many staff, while they reveal only emerging formal structures and mechanisms for communication. Meanwhile, the innovation that was crucial to the inception and early success of the company continues. Interviews also suggest that the company is struggling to achieve a greater degree of formalization of rules, a more stable structure, and improved efficiencies. Therefore the company appears to be grappling with aspects of the formalization stage without having entered such a phase. The results of the Learning Organization Survey (Garvin et al., 2008), seen in the context of Daft’s (2010) life cycle model, reinforce the above conclusion. As described in the Chapter 3 subsection Learning Organization Survey, the first building block of the survey, Learning Environment, includes questions grouped under the following components: Psychological Safety, Appreciation of Differences, Openness to New Ideas, and Time for Reflection. The items measured do not require formal systems and structures. This building block is therefore compatible with an organization in stage one (entrepreneurial), or stage two (collectivity) of its development according to Daft’s model. The second building block, Concrete Processes and Practices, on the other hand (which includes the components of Experimentation, Information Transfer, Analysis, Education and Training, and Information Transfer), requires more formalized

164

systems and implementation. The results of the survey therefore provide further evidence that the company is in the collectivity stage of its development because they indicate strengths in the first building block while suggesting significant weaknesses in the second. The above conclusion implies that in order to more completely establish the conditions for the development of a learning organization, the company must progress from stage two (collectivity) to stage three (formalization) in its development. While levels of organizational learning happen automatically as people work together through processes of trial and error, informal communication, and interpreting explicit and implicit feedback from co-workers and superiors, more deliberate and systematic levels of organizational learning require more formalized structures. Furthermore, if the company wishes to engage in deliberate and selective knowledge management practices, this also implies a stage of development which embraces a more conscious and strategic approach to knowledge than what is likely to be found in the collectivity stage. The question of whether the company in this study will necessarily move from stage two to stage three, however, is more complex because it depends on other factors. Despite its early exponential growth, the company remains quite small—employing 115 people. As stated earlier, Greiner (1972) argues that one of the key factors driving stages of development is the size of the company; and Quinn and Cameron (1983) assert that the power of constituencies to change criteria for organizational effectiveness is another key factor. According to these two theories, therefore, the company won’t enter a formalization stage until the number of employees and the scope of its operations increases to the point where they trigger a management and leadership crisis; and/or unless the founders lose control of the company to constituencies such as investors

165

who impose organizational effectiveness criteria which strongly emphasize formalization and control. Nothing in the current situation of the company suggests this is imminent. In a fast evolving sector such as e-commerce, companies must simultaneously exploit their current advantages while exploring new opportunities becoming what Grant (2010) refers to as “innovating organizations”, and practicing what Hitt, Ireland, Sirmon, and Trahms (2011) refer to as strategic entrepreneurship. According to Grant, organizations which continuously seek to innovate should tend towards flat hierarchical structures and task-oriented project teams. The processes of such organizations should emphasize loose controls, idea generation, and flexible strategic planning; their reward systems should be based on autonomy, recognition, and equity participation in new ventures. Finally, people in such organizations need to combine technical knowledge and creativity. This description seems to fit the study company well; moreover, Grant’s theory suggests that, in order to remain innovative, the study company should maintain many of these characteristics, which appear incompatible with the formalization stage. Furthermore, Rowden’s (2001) version of the learning organization as a model for strategic change emphasizes that the company must be constantly ready to embrace change, continuously planning, improvising execution and encouraging experimentation, not to mention engaged in action learning (p. 15-16). Perhaps by strategic design, and/or thanks to the ability of the founders to maintain control, the company will not evolve to a formalization stage of development as per Daft’s (2010) model, but will remain a leaner more flexible entity for the foreseeable future. If that is the case, the company will need to adapt aspects of the formalization stage sufficient to allow practices and processes to develop which are required for a learning organization, according to the second building block in the Learning Organization Survey

166

(Garvin et al., 2008). Hence a larger question related to this study, and one that can only be answered by further research, is: can the learning and knowledge life cycle of the study company progress towards learning organization and knowledge management practices without the company entering a true formalization stage of development—i.e. without experiencing a leadership and management crisis? An additional and related question that arises is: if the company resists moving towards a true formalization stage, will that come with costs such as unrealized potential growth, and stagnating or declining efficiency? Readiness for change in the study company. In Chapter 4, in the subsection Internal comparison, the results from the survey were compared across departments and across levels of hierarchy; they revealed some important divergences. While the sample sizes were too small to meet the accepted criteria for statistical analysis of differences, the practical and theoretical implications of these divergences are briefly discussed in the following paragraphs. First, Table 10 (survey results across departments) shows that the human resources department has the largest number of highest per item scores—seven out of eleven—including for both the Learning Environment Composite, and the Learning Process Composite. Meanwhile, the e-commerce department shows the largest number of per item lowest scores— nine out of eleven—also for both the Learning Environment Composite, and the Learning Process Composite. Hence these two departments appear to be on nearly opposite ends of the spectrum when it comes to the survey results; meanwhile, the other departments fall in between without any apparent pattern. Second, Table 11 (survey results across levels of hierarchy) suggests another interesting contrast, even if the overall degree of variation is lower than the comparison across departments. The executive committee has the lowest average scores for all the items in the first building

167

block, including the Learning Environment Composite; while the managers show the highest scores for all the same items. The pattern switches in the second building block, where the executive committee shows the highest scores for all but one item (Information Collection), and the employees show the lowest scores for all but one item (also Information Collection). The results of these comparisons among departments, and across levels of hierarchy, demonstrate a lack of homogeneity in the employee’s perception of the company’s learning environment and learning practices. Such differences may reveal different degrees of readiness for change as per Rowden’s (2001) strategic learning organization model; i.e. they may reveal differences in the extent to which certain departments and levels of hierarchy consider change to be necessary. For example, if the average survey score for the executive committee for Education and Training is 67.24 compared to 48.71 for the employees, we may conclude that the employees perceive a greater need for Education and Training than the executive committee. The practical implications, of course, are significant here because the executive committee holds the greatest decision making power and access to resources; it may need more convincing when it comes to the necessity for training than the managers or employees, who, on the contrary, appear eager to embrace it—the topic of power and organizational learning is dealt with in more detail below. To take another example, if the company were to prepare to introduce new or reinforced learning processes and practices (second building block), they should consider, as a first step, how to address the e-commerce department’s perception of a weak learning environment (first building block). Moreover, the e-commerce department’s average score for Time for Reflection was 37.31 compared to 53.17 for operations, 60.05 for commercial/marketplace, and 82.34 for human resources. This suggests that e-commerce employees perceive the time pressures

168

associated with their work activities to be excessively high; such a perception works against a favorable learning environment. Considering the e-commerce department is the largest in the company, these discrepancies are worthy of consideration. Cycles of knowledge creation and learning processes: how knowledge gets generated and disseminated within the organization. Dynamic models of knowledge creation and organizational learning cycles can provide further insights to this discussion. Nonaka’s (1994) modes of knowledge creation, for example, which was briefly described in Chapter 2 in the subsection Knowledge management, suggests how individual tacit knowledge may be shared among members of the organization to be rendered partially explicit, then combined with existing knowledge, and finally assimilated on a company-wide level through experimentation and practice. It is a cyclical model whereby the last mode, assimilation of explicit knowledge, feeds into the beginning of the cycle when the explicit knowledge becomes embedded in practices and routines, and thus becomes tacit over time. If we consider the survey results in the context of this model, one possible interpretation of the comparatively low e-commerce scores is that the tacit knowledge possessed by e-commerce employees is not shared effectively due to lack of socialization opportunities; hence its potential for generating new organizational knowledge may be underexploited. A possible interpretation of the comparatively low executive committee scores for the second building block is that there is a lack of systematic experimentation and institutionalized implementation of new knowledge to create opportunities for assimilation. Only further research could help determined if these interpretations are accurate. An additional framework which holds insights for analyzing the study results, is the organizational learning framework developed by Crossan et al. (1999)—also briefly described in

169

the Chapter 2 subsection Knowledge management. This framework presents a dynamic strategic renewal model for the flow of learning in an organization based on four key processes which account for how learning moves up from the individual, to the group, to the institution, and then back down towards the individual again. These key processes, referred to as the 4Is, are: intuiting, interpreting, integrating, and institutionalizing. Each of the four learning processes is briefly described below. The process of intuiting is highly individual, tacit, and gained through experience; it is associated with the development of past-oriented expertise, on the one hand, and future-oriented entrepreneurial intuition, on the other. Intuiting can be defined as the “preconscious recognition of the pattern and/or possibilities inherent in a personal stream of experience” (Weick, 1995, as cited in Crossan et al., 1999, p. 525). The process of interpreting refers to when individuals use words or actions to explain to themselves and/or to others hunches, feelings, insights and ideas (Crossan et al., 1999). Links can be made here to Ambrosini and Bowman’s (2001) notion of tacit knowledge which can be expressed imperfectly through the use of language. This process links the individual to the group because it allows for the transmission of knowledge and experience among individuals; furthermore, it facilitates intersubjectivity, and the resolution of equivocal situations through discussion. The third process in the framework developed by Crossan et al. (1999), integrating, is when individuals develop a shared understanding of new knowledge, an insight, or an innovation, which leads them to take coordinated, collective action. This process is associated with work groups, teams, and communities of practice. Finally, the fourth process, institutionalization occurs when the learning gets “embedded in the systems, structures, strategy, routines, prescribed practices of the organization, and investments in information systems and

170

infrastructure” (Crossan et al., 1999, p. 528). Similar to Nonaka’s (1994) model, therefore, the explicit learning becomes tacit as the conscious, deliberate learning processes are embedded into daily practices. Table 12. Tacit skills revealed during semi-structured interviews analyzed according to Crossan et al.’s (1999) learning processes Key organizational constructs from semi-structured interviews Organizational learning processes Intuiting (individual-level)

Interpreting (individual-level or among several individuals)

Integrating (group-level)

Institutionalizing (organizationlevel)

Innovation

Knowledge gained through experience about the sector, knowledge of brands. Anticipating trends, thinking creatively to exploit the trends/opportunities. Discussing how to turn ideas and “vision” into concrete reality. Discussing how to exploit tends and opportunities.

Commercial success Communication skills – how to be credible, charismatic, etc. when dealing with the client.

Exchanges among employees about what works and what doesn’t, sharing experience.

Getting people to work together in groups to realize projects, offer services, find clients.

Listening to the client and making a customized offer. Implementing a sales strategy, building commercial teams.

Implementing departments and company policies. Development of an ecommerce platform

Rubrics used by commercial team to analyze client needs.

Good relations with clients Knowledge gained through experience in the sector, knowledge acquired about brands. Specific knowledge of the individual client’s priorities, corporate culture. Communication within teams/departments and across teams/departments.

Effective teamwork skills like flexibility. Cooperation across different departments to deliver the sites and services. Management of work flows and flow of information. Training new employees.

Client and/or employee development Understanding the client’s point of view. Knowing what information should be shared with the client.

Honing communication skills thanks to feedback – how to be credible, charismatic, etc. when dealing with the client. Management practices such as asking the right questions and delegating. Employee training, rubrics and criteria used during interview/hiring process.

Table 12 presents skills revealed during the semi-structured interviews and organizes them in rows according to the four learning processes developed by Crossan et al. (1999), and in columns according to the key organizational constructs—described in Chapter 4 in the subsection Causal mapping analysis—with which they are associated. The table allows connections to be made across the skills revealed, the organizational success factors, the four

171

learning processes, and the levels at which the learning takes place. For example, concerning the key organizational success construct, innovation, intuiting involves the individual insights and vision the founders had as they were starting the company; these were then interpreted (i.e. articulated and discussed), before being integrated (i.e. developed into coherent and coordinated actions), and then finally institutionalized (i.e. implemented into the company structure and policies). Seen in this way, the learning at the company transpires in a continuous cycle and involves a flow of tacit skills to explicit skills and back to tacit skills. While Crossan et al. state that intuiting is purely individual, compared to institutionalization, which is purely collective, they conjecture there is natural overlap between interpreting and integrating. For instance, concerning the key organizational success construct, commercial success, the interpreting processes, which include discussing effective practices among colleagues, and listening to the client, would not be easy to separate, in reality, from the integrating processes of developing the sales strategy. Lawrence, Mauws, Dyck, and Kleysen (2005) make an important contribution to Crossan et al. (1999) by integrating notions of politics and power into the organizational learning framework. In their view, the framework developed by Crossan et al. (1999) does not adequately account for why only some useful innovations are embraced by organizations while others languish; nor does it account for why some organizations are better at learning than others. Lawrence et al. conjecture that power plays a key role in explaining this, and they outline different political strategies for exercising power in the organization which most effectively facilitate each of the four learning processes. These strategies are briefly described in the following paragraph.

172

Lawrence et al. (2005) identify two modes of power: episodic and systemic. The former type of power refers to “discrete, strategic political acts initiated by self-interested actors” (p. 182), whereas the latter refers to power which works “through the routine, ongoing practices of organization” (p. 182). Lawrence et al. maintain that episodic power is exercised by individuals through strategies such as influence (persuasion and negotiation), or force (agenda setting, limiting alternatives, firing employees); while systemic power is exercised by organizations through strategies such as domination (design of material technologies, information systems, and physical layout), or discipline (socialization, training, team-based work). According to this model, in order for the process of intuition to yield useful innovations, it should be facilitated by a political strategy based on discipline so that individuals acquire rich and meaningful job experiences, and identify with the organization. In order for new ideas to gain acceptance through the process of interpretation, a political strategy based on influence requiring strong social skills and relevant expertise will be most effective. In order to most effectively facilitate the process of integration of new ideas into group activities, a political strategy that involves the use of “force based on formal authority” (p. 186) is advised. Finally, in order for the process of institutionalization to effectively lead to the adoption of innovations on a company-wide basis, a strategy of domination (i.e. embedding the innovation in the institutional design, technologies, and/or company layout, etc.) is recommended. The model developed by Lawrence et al. (2005) has pertinent theoretical and practical implications for interpreting the results of both the survey and the semi-structured interviews. For example, it suggests that if the company wishes to encourage useful innovations on an individual level, investments in education and training, and developing good team work practices would be most effective. While this may appear counterintuitive because education, training and

173

teamwork tend to have a collective focus, this strategy will help foster a sense of identification among individual employees with the company, and it will develop a deep understanding among staff of what the company expects of its employees. This model furthermore implies that the interpreting process, where ideas gain acceptance, is difficult to control because it depends largely on unique, individual abilities to influence co-workers and colleagues, such as social skills and charisma, as much as on formal positions in the organization. Not imposing rigid controls to manage this process may in fact be preferable than the contrary in order to allow potential intuitions to emerge from anyone in the company. The model developed by Lawrence et al. also implies that for the company to effectively integrate new ideas into group activities for realizing coherent collective action, people with legitimate and real power (i.e. top management) will have to back the ideas using their formal authority as necessary to make sure the innovations are implemented. Finally, the model suggests that if the company wishes to implement innovations on a company-wide basis, the most effective political strategy would be to embed them in the technology which employees use, and/or in the organizational structure, policies, and/or in the floor plan.

174

Discipline

Institutionalizing

Intuiting

Domination

Influence

Integrating

Interpreting

Force

Figure 20. Integrating political strategies into the 4I learning processes framework (adapted from Lawrence et al., 2005) Figure 20 presents the four learning processes proposed by Crossan et al. (1999) in their 4I framework, and the political strategies outlined by Lawrence et al. (2005) in a circular knowledge cycle. The diagram shows where in the cycle the political strategies can be the most effective at disseminating useful innovations.

175

Internalization

Institutionalizing

Intuiting

Socialization

Externalization

Integrating

Interpreting

Combination

Figure 21. Knowledge generation, and learning processes (adapted from Crossan et al., 1999 and Nonaka, 1994) Figure 21 combines the learning process framework developed by Crossan et al. (1999) with Nonaka’s four modes of knowledge creation. The diagram suggests that, while the two models have distinct areas of focus—learning processes for Crossan et al. (1999), and new product development for Nonaka (1994)—they are very compatible, and can even be considered as describing different elements of the same knowledge and learning cycle. Socialization is the knowledge management tool best suited for moving learning from the process of intuiting to the process of interpreting; combination is the mode which can move learning processes from interpreting to integrating, et cetera.

176

•Explicit skills

•Deeply ingrained tacit skills

Institutionalizing

Intuiting

Integrating

Interpreting

•Tacit skills that could be articulated

•Tacit skills that can be imperfectly articulated

Figure 22. Learning processes and degrees of tacitness (adapted from Ambrosini & Bowman, 2001 and Crossan et al., 1999) Once again, Figure 22 presents the four learning processes proposed by Crossan et al. (1999), but this time adds the degrees of tacitness proposed by Ambrosini and Bowman (2001) which form the theoretical foundation for operationalizing tacit skills in the causal mapping method. Figure 22 suggests that, as learning processes move in the cycle from intuiting to interpreting to integrating, and finally to institutionalizing, the skills involved become progressively more explicit. The cyclical nature of the flow of knowledge and learning, not to mention the interplay between tacit and explicit skills depicted in Figures 20, 21 and 22, account for some of the inherent challenges studying these phenomena. It is easier for employees to talk about explicit skills and learning which have been fully implemented than it is to talk about skills which are partly unconscious and hard to articulate; it is moreover difficult for employees to talk about learning processes located at the intersection of tacit and explicit skills—learning processes

177

which are evolving and full of ambiguity. What is more, these challenges may partly explain why the knowledge-in-practice framework (McIver et al., 2013) failed to reveal many tacit skills associated with specific positions. These challenges are discussed more in the next subsection on Measurement. Parallels can be drawn between the above model and Daft’s life cycle stages of development (2010) discussed earlier. If the company is in stage one (entrepreneurial), or stage two (collectivity) of the lifecycle, the learning and innovations will be largely spontaneous, and people-driven because they arise out of individual intuition without company-managed education and training. Moreover, the adoption of such knowledge and innovation will depend on individual influence (i.e. charisma and forceful personalities), and the legitimate authority of those people with the power to make sure innovations get implemented. These types of power correspond with an upward flow of learning and innovation—i.e. from individuals to collective and institutional levels. Crossan et al. (1999) refer to the directions in the flow of learning using the terms feedforward (instead of up), and feedback (instead of down); they posit that the former is associated with the exploration and generation of new knowledge and ideas, whereas the latter with the exploitation of knowledge and learning already possessed by the company. Elements of stage three (formalization), as the company matures, will probably correspond to the use of systemic power on an institutional level to push learning and innovations back down towards individual learning. The theme of knowledge dissemination within the organization is dealt with further below in the discussion about team work and team learning. The role of tacit knowledge and skills in the development of organizational culture. One of the reasons the company cooperated with this study is because certain members of the executive and managerial leadership were interested in developing a learning organization

178

culture in line with its business objectives and modus operandi. The development of learning in an organization is closely related to the development of an organizational culture. Schein (2010) maintains that organizational culture is based on a pattern of shared assumptions; according to this definition, culture is a product of social learning which develops over time. Meanwhile, Hofstede (2001) contends that “shared perceptions of daily practices” (p. 394) form the core of an organization’s culture. Other scholars emphasize shared values as the basis for corporate culture (Peters & Waterman, 1982). Both Schein’s and Hofstede’s definitions are clearly relevant to this study, as well as to the concept of organizational learning, in general; Schein’s definition is particularly germane, hence it is quoted in full below: The culture of a group can … be defined as a pattern of shared basic assumptions learned by a group as it solved its problems of external adaptation and internal integration, which has worked well enough to be considered valid and, therefore, to be taught to new members as the correct way to perceive, think, and feel in relation to those problems. (Schein, 2010, p. 18) According to Schein’s (2010) concept of organizational culture, the most significant level of culture consists of deeply held assumptions which are learned over time thanks to experience solving problems; these assumptions are difficult to detect for a researcher but greatly influence and help explain the more easily visible levels of culture which are artifacts (e.g. observable behavior, or a company floor plan) and espoused beliefs and values (e.g. what people say they believe). Revealing the deeply held assumptions presents similar challenges to revealing tacit knowledge. Assumptions are largely unconscious and un-discussable (Schein, 2010); they can only be deducted by a researcher thanks to observation and/or effective interview techniques. Tacit skills and knowledge are likely to be intertwined with these assumptions. Hence, the semistructured interview data gathered for this study undoubtedly holds useful information about the company’s organizational culture, according to Schein’s definition, even if that wasn’t the explicit intention of the interviews. In a similar vein, the survey results gathered for this study 179

reveal the employees’ perceptions about the learning environment and practices, and thus also reveal elements of the organizational culture according to Hofstede’s (2001) definition. A complete assessment of the study company’s organizational culture, however, is beyond the scope of this dissertation. Not only is it a common sense conclusion that understanding the culture of an organization requires an assessment of the national cultural context, but scholars also emphasize that an effective appraisal of corporate culture requires an assessment of the larger cultural context including national and occupational culture (Hofstede, 2001; Schein, 2010). A thorough assessment of the company culture in this study would therefore require an analysis of the influence of French national culture, core technologies used daily by the employees, and the occupational cultures of some of the most represented jobs (e.g. web developers, traffic analysts, et cetera). Moreover, revealing underlying assumptions is complex and time consuming; Schein argues that an adequate understanding of group culture entails an assessment of several major underlying assumptions, which may even appear contradictory, but which operate simultaneously in a dynamic fashion; a researcher needs to grasp all of these major underlying assumptions and how they function together in order to really understand the meaning of the whole, or of any one of the assumptions. Lastly, organizational culture, according to Schein, can only develop over time as the members of the organization face and solve problems; espoused values of company representatives do not become true underlying assumptions until they have been tested by challenging circumstances, problems, or crises. Because the personnel at the study company has changed so much in its relatively recent history—with staff turnover sometimes reaching 40% or more—it is difficult to assert with confidence that the company has developed a real organizational culture that goes much beyond

180

artifacts, and espoused values. Nevertheless, what follows is a brief analysis of the elements of the organizational culture deducted from the study findings. Some of the artifacts I observed during my interviews in the company include the loftlike open architecture floor plan—at the time the company was located in a former parking garage that had been converted into an office space—with functional furniture; a dynamic fastpaced environment; a well-stocked refrigerator, and break rooms with sofas and games. There was a sense of egalitarianism because everyone dressed informally, and because it was impossible to determine from appearances any hierarchy. One of the recurrent espoused values, which came up particularly during the individual interviews with the founder and the human resources manager, was the importance of employee autonomy—i.e. employees are expected to take the initiative and question the advice and opinions of colleagues. Anyone reading this passage who is familiar with Schein’s (2004, 2010) work on Digital Equipment Corporation might recognize similarities to the artifacts and espoused values he describes encountering in that company. One possible interpretation for this is that because both companies deal with electronics technology and are (or were) on the cutting edge of business technology trends, the similarities stem from similar occupational cultures—even if the companies are not in exactly the same activity. Another possible interpretation is that the study company’s founders have been influenced by other companies’ examples of corporate culture, such as DEC, and have consciously imitated the culture which they associate with innovation and technology. This second interpretation is reinforced by the follow up interview which was conducted with the e-commerce manager and the human resources manager. During this interview, the human resources manager questioned the classification of some of the skills I had grouped under the organizational success construct, innovation, as being tacit; in her view, some

181

of the management practices I assumed had been learnt through trial and error, had in fact been learned by the founders and managers reading books about technology startups—i.e. an explicit transfer of knowledge and ideas. At any rate, further research would be required to determine if the artifacts and espoused values I observed go beyond appearances, and have actually developed into deeply held assumptions and widely shared perceptions of daily practices sufficient to form a meaningful organizational culture. The development of a true organizational culture, because it involves developing a consensus around issues of internal integration and external adaptation, would be closely related to the development learning processes and practices, and closely related to the organization’s life cycle stages of development. Role of tacit information in the development of successful teamwork. The work of Edmondson (1999, 2002), one of the co-authors with Garvin and Gino of “Is Yours a Learning Organization” (Garvin et al., 2008), is particularly relevant not only to this study, but also to our general understanding of how teams function and how they learn; her concept of Psychological Safety forms one of the key components of the Learning Organization Survey (Garvin et al., 2008). According to Edmondson (2002), organizational learning transpires thanks to interactions between individuals who are part of a small group or team within the larger organization; she states that “collective learning is inherently local” (p. 142). This position reinforces Crossan et al. (1999) who, as seen above, conjecture that organizational learning processes such as interpretation and integration are likely to involve small groups and teams. Work teams can be defined as: “groups that exist within the context of a larger organization, have clearly defined membership, and share responsibility for a team product or service” (Edmonson, 1999, p. 351, referring to the prior work of Alderfer, 1987; and Hackman, 1987). Teams are the most fundamental unit in organizational learning and a necessary vehicle

182

to move individual knowledge and expertise of a tacit nature to its wider utilization within the organization—i.e. from individual/tacit to organization-wide explicit (Edmondson, 2002; Smits, personal communication, 2017). Understanding how teams learn and what factors influence the learning process is a vital question to address, and it is linked to the issue of how the organization produces new ideas and capabilities (Edmondson, 2002). While the research for this study did not directly focus on teams, the data from the survey and the semi-structured interviews reveal useful information about the role of small groups and teams in the company and the potential for learning which can be found there. Applied to the study company, Edmondson’s (2002) definition of work teams covers a range of small group structures including the executive committee, the sales team, project-based teams, and client-based teams. Some of these groups are relatively permanent, in so far as they meet regularly and continuously, while others last only as long as the project to which they are assigned. Some of the teams have a broad scope of decision making authority while others are more narrowly focused on the execution of tasks. The following paragraph outlines the different groups found in the company that can be considered work teams. The executive committee comprises seven people including the president/co-founder, the financial director, the head of commercial development and strategy, the e-commerce manager, the operations manager, the head of marketplace, and the head of photography/co-founder. Each member has very distinct areas of responsibility and expertise. They do not necessarily work closely together on a daily basis but meet approximately once a week to deal with high-level operational and strategic issues. Also, there is the sales team which comprises the head of commercial development and strategy, and three employees who report directly to him. This team works together regularly with one overarching goal: to develop the company’s client base.

183

There appear to be several project-based teams in the company. One such team, which was discussed during the group interviews, was created to develop the proprietary e-commerce platform which the company hopes to license in the future to clients. This team, comprising eight to nine technical employees, is thus a research and development initiative and will presumably last as long as it takes to bring the project to fruition. Meanwhile, other project teams are created each time the company signs a new contract to build the e-commerce websites according to the client’s specifications. Members of these teams report directly to a project manager who is attributed to the project once a new contract is signed; the number of members was not specified in the interviews but it is presumably small (four to five) comprising only the necessary technical and marketing skills to build the site. Finally, operational teams exist for each client brand; these teams are referred to as “e-stores” and include two e-store managers (one junior, one senior) plus a small number of technical employees. The e-store teams report to a brand manager and may include some of the members of the project team that initially created the client website. The members of these teams are in constant contact with one another, and with the client, to manage the sites; and to deal with issues pertaining to product display, brand image, promotions, logistics, inventories, and any technical problems which arise. The e-store teams are part of the management and operational structure of the company. Now that the different types of teams in the study company have been briefly sketched out, the concept of team learning will be appraised in the context of the enterprise. Team learning can be defined as “a process in which a team takes action, obtains and reflects upon feedback, and makes changes to adapt or improve” (Edmondson, 2002, p. 129, citing the earlier work of Edmondson, 1999; and Argote et al., 2001). This definition is close to the conceptualization of group learning discussed earlier in this paper (Wilson et al., 2007)—the

184

main difference is the latter’s insistence on fundamental processes of sharing, storage and retrieval. A team may have the opportunity to engage in double-loop learning (Argyris, 2002; Argyris & Schon, 1996)—what Edmondson refers to as “radical learning” (2002)—to the extent that its range of decision-making authority and action allow it to. The executive committee, for example, would likely have the opportunity for this type of learning because of the complex and ambiguous nature of the issues the members would face, the lack of clear, available information, and because of the inherent risk taking. On the opposite end of the spectrum, the e-store teams would more likely only have the opportunity to engage in single-loop learning (Argyris, 2002; Argyris & Schon, 1996)—what Edmondson refers to as “incremental learning”—because their scope of action and decision making would likely be limited to making improvements in the efficiency and effectiveness of their work rather than allowing for a broader reflection about the choice, and meaning of those actions. Having observed several teams in action, Edmondson (2002) concludes that team learning is variegated even when teams appear to have much in common; i.e. when team composition, missions, levels of hierarchy represented are all quite similar, some teams will nevertheless engage in learning processes more effectively than others. Edmondson’s concept of Psychological Safety (1999, 2002) developed out of her observations that when team members do not feel comfortable sharing different points of view, or talking about errors for fear of retribution and/or of not being taken seriously, the team’s ability and willingness to talk openly, not to mention experiment with new ideas is negatively impacted. A precise definition of Psychological Safety is provided below: The term is meant to suggest neither a careless sense of permissiveness, nor an unrelentingly positive affect but, rather, a sense of confidence that the team will not embarrass, reject, or punish someone for speaking up. This confidence stems from mutual respect and trust among team members. (Edmonson, 2002, p. 354)

185

Edmondson (2002) conjectures that inappropriate deference to authority negatively impacts a team’s ability to engage in productive reflection about work activities. Other observed impediments to team learning include ignoring relevant information, and failing to implement decisions. Edmondson further hypothesizes that failure to implement decisions itself stems from multiple possible causes such as lack of resources, insufficient motivation, and the inability to break out of comfortable routines. On the other hand, group norms which encourage working interdependently and encourage open communication positively impact team learning (Edmondson, 2002). The group interview provides data about how employees learn from one another, how teams function, as well as about how communication transpires across teams and departments. The data suggest that learning in the company is highly social, that employees are keenly aware of how important this is, and that informal social exchanges are the principal means by which several employees learned how to do their jobs. Margaux, for example, a supply chain logistics manager in Group Interview 2, states that by exchanging with others, she is able to grasp the core concerns of her job rather than “drown” in the overwhelming amount of details. Also in Group Interview 2, Raymond, head of traffic and CRM, contends that high performance and success are not individual accomplishments, and that it’s necessary to have “empathy, comprehension, and sharing” to achieve high performance. Then, in Group Interview 4, Maud, a project manager, explains that when she was hired, her manager was on paternity leave—exchanging with others was one of the main ways she learned on the job. Also in Group Interview 4, Madeleine who is an account manager in Marketplace (but who started as an assistant e-store manager), states that her manager left soon after she started—consequently, she also learned on the job with little formal training thanks to exchanging with colleagues.

186

In addition to demonstrating the importance of social exchanges, the interview data provide evidence for an environment favorable to team learning. For example, when Raymond talks about the importance of having a global understanding of the activity in order to effectively solve problems, Margaux responds to this by referring to a “state of mind” which involves the willingness to take initiatives, and not being afraid to fail. When three of the other four employees in Group Interview 2 agree, this validates the survey results concerning Psychological Safety which appears to be quite well established in the company. Moreover, when Mariane, an assistant e-store manager in Group Interview 3, defines success for her job, like Raymond, she emphasizes how much her success depends on others including the client, the web designer, and the web integrator. In a similar vein, in Group Interview 5, Jocelyne, an e-store manager, when asked who she speaks to when she has a problem, states that she speaks to her e-store partner (presumably about business related issues), or to Geoffroy, the lead IT architect, if the issue concerns IT migration issues. As seen above, group norms which encourage interdependence are a positive factor in team learning. To sum up what the employees seem to be learning from one another in teams and small groups, we can note the following: elements of the company culture (the way of working in the company), how to prioritize amid an overwhelming amount of information, and how to solve problems by thinking holistically. Despite evidence of robust social exchanges among employees, evidence of group norms encouraging interdependency and other elements of Psychological Safety, the interview data also reveal areas where tacit individual knowledge does not appear to successfully move to a wider utilization in the organization, and/or where teams function unevenly. As suggested by the survey data discussed in the Chapter 4 section Survey Results, the company culture seems to favor informal social exchanges rather than more formalized processes, and this creates difficulty

187

not only in terms of storing and retrieving knowledge but also in managing the flow of information across teams and departments. Consistent with Edmondson’s (2002) theory, team behavior in the company appears variegated—hence, the team learning is also likely to follow suit. In Group Interview 1, Rodrigo, a web developer, maintains that forming teams is difficult, and he cites high turnover of team members as a factor. Moreover, although in Group Interview 1, Kristina, who works in finance, describes team behavior where everyone listens to everyone else, “even the intern”, Adrien, who conducts search engine optimization, states that “not all teams function well”, and he cites a lack of action to follow up words. Also, although in Group Interview 4, Solange, who works in human resources, describes weekly team meetings and appears impressed at how much teamwork counts in her job compared to her prior job (in another company), during Group Interview 3, Benoit, a junior e-store manager, laments that his team only meets when there are problems. Benoit is also the person who uses the term “free style”, noted in the Chapter 4 section Answers to Research Questions, as a way to describe when the team must work so quickly to deliver website features that they deliberately ignore certain processes—this was noted in the relevant section as potentially being an example of counterproductive learning, and it is discussed in more detail below. Likewise, a negative example, previously noted in the Chapter 4 section Answers to Research Questions, concerns a position that was conceived to lead technological innovation but was discontinued after a year. Adrien’s explanation for this in Group Interview 1 is that the position was discontinued due to a lack of “tangible result”. Although this example suggests possible tension between short-term commercial objectives and longer-term investments in innovation, it may also illustrate a failed approach that relied too much on an individual position. Meanwhile, an example from Group Interview 2 (also mentioned briefly in the Chapter 4

188

subsection Causal mapping analysis) was provided by Geoffroy, the lead IT architect, when he describes the resistance he encountered trying to implement a new source code. He first had to “sell it” (i.e. the source code) to his manager, who then had to “sell it” to the marketing team, who then had to “sell it” to the executive committee. This example can be seen in light of Crossan et al. (1999) who suggest that innovations, when they stem from individual intuition, must be interpreted by other members of the organization before being integrated into practices and processes—i.e. before moving up to wider utilization, and that this process depends largely on the influence of individuals. Another telling example in Group Interview 1 comes from Christian, a web developer, who explains that one employee attempted to develop a manual for setting up good websites, but that only he (the employee) understood it; hence the company still lacks an effective manual. If we add to this the example given by Adrien about the person in charge of payment systems who resigned thus leaving the team without adequate knowledge of how to set up such systems (discussed previously in Chapter 4 in the subsection Causal mapping analysis) we have several illustrations which suggest that individual knowledge is often not effectively moving to a wider utilization in the organization, and that this problem may need to be a addressed on the level of small groups and teams. This problem is likely to impact the flow of information and coordination across teams which, as noted in the relevant Chapter 4 subsection, is described as problematic in the interviews by several employees—for example, by Christina and Eleonore in Group Interview 1. Crossan et al. (1999) posit that teams are not only crucial for the flow of learning from individuals to a wider organizational utilization (feedforward) but also for the implementation of processes and innovations which move from the organizational level back down to individual behavior (feedback).

189

Main conclusions drawn from the theoretical discussion. To sum up this discussion of theory, despite the unique characteristics of e-commerce and the lack of research in the sector dealing with life cycle stages of organizational development, this body of theory can provide insights about the study company, and can help frame some of the study findings. The company fits many of the characteristics of the collectivity stage; however, it is difficult to speculate whether the study company will or even should transition to a formalization stage (Daft, 2010). Even if becoming a strategic learning organization would seem to require more formalization of certain processes and practices than are currently in place, the need to remain agile in order to quickly adapt to and to learn from the external environment has to be weighed against the possible downsides of formalization. The lack of homogeneity across departments and levels of hierarchy in the survey results for specific items suggests different levels of readiness for change within the company (Rowden, 2001). The implications are that building a stronger perception of a favorable learning environment among e-commerce employees would be advisable before developing more formalized learning processes and practices; and also, that the executive committee may require more convincing when it comes to the need for training, than the managers and employees. Theories of knowledge generation and learning processes can likewise provide useful insights about the study findings and they have implications regarding the how the company can reinforce learning practices in order to cultivate useful innovations. In addition, links can be made between theories dealing with life cycle stages of company development (Daft, 2010), cycles of knowledge generation (Nonaka, 1994), cycles of learning processes (Crossan et al., 1999), and degrees of tacitness of skills (Ambrosini & Bowman, 2001). Figures 20-22 combine

190

elements of the above theoretical perspectives to render the links between the different theories easier to visualize. Although doing an in-depth analysis of organizational culture of the company is out of the scope of this study, and even if it is difficult to argue with confidence that such a young company, which has experienced relatively high staff turnover, has even developed a true organizational culture, as per Schein’s definition (2010), it is possible to sketch a brief outline of observations and cultural artifacts revealed in the study data. These observations and artifacts suggest an organizational culture similar to other cutting edge technology companies in the electronics sectors (notably Digital Equipment Corporation). These similarities may be explained by similarities in roughly aligned occupational cultures, but they may also be the result of conscious imitation on the part of the founders who were inspired by reading books about technology startups. Regarding theories of tacit learning and teamwork, relevant literature suggests that work teams are key for disseminating tacit knowledge (Crossan et al., 1999; Edmondson, 2002); several examples and different types of work teams can be identified in the study company. Teams can positively influence feedforward learning processes, as well as feedback learning processes (Crossan et al., 1999). The study data reveal some factors which can facilitate effective team work, such as interdependency and Psychological Safety. However, the data also suggest that team learning is variegated with some teams functioning better than others. Moreover, the data reveal instances and situations where tacit information is not getting disseminated effectively, as well as revealing some examples of counterproductive informal learning.

191

Measurement. Utility of causal mapping in mixed methods research. This subsection will assess the strengths and weaknesses of the causal mapping method in mixed methods research. Before assessing those strengths and weaknesses, however, the inherent challenges to studying tacit knowledge will be reviewed because those challenges should not be confused with weaknesses in the methodology. As discussed in the Chapter 3 subsection Semi-structured interview protocol, tacit knowledge is at once vast and ambiguous while being personal, highly context specific, and rooted in practice. There is intrinsic tension therefore when the researcher attempts to capture clear, generalizable elements of tacit knowledge from the details of daily practices and routines of employees; ultimately this tacit knowledge cannot be completely detached from its context nor clearly communicated through words, and this is exactly what makes it potentially valuable; it is firm-specific and cannot be imitated. Hence the partial and fragmented nature of the causal mapping method, noted by Ambrosini and Bowman (2001), rather than being a methodological weakness, is more accurately described as an inevitable limitation of any attempt to capture tacit knowledge. Because of the inherent challenges described above, the skills revealed by the causal mapping analysis (described in the Chapter 4 section Results from Semi-structured Interviews) may appear imprecise. Some examples include the ability to work in a dynamic, fluid work context with loosely defined processes; the ability to prioritize under stress and to deal with uncertainty; and teamwork skills such as flexibility in problem solving, and the ability to share responsibility. These skills read more like vague clusters of abilities or even personality traits rather than clearly delineated skills like we might expect when describing explicit, technical competencies such as the mastery of specific software programs or pieces of medical technology.

192

This raises some difficult questions. What does a tacit skill really consist of? What links the ability to prioritize under stress in two very different job functions? And where does the boundary lie between personality traits, and skills when dealing with a concept like flexibility? The following paragraphs attempt to address these questions. It was already briefly mentioned in the Chapter 4 subsection Causal mapping analysis that flexibility should be considered a skill, rather than a trait, to the extent that it is linked to work practices and can be developed over time. In order to clearly distinguish between the terms “soft skills” and “personality traits”, Matteson et al. (2016) argue that traits are individual attributes which remain relatively stable over time, and affect behavior, while skills “inherently involve performance, action, or change” (p. 74). According to that distinction, an employee could be taught or trained how to be flexible in a specific work setting for the purposes of realizing work-based outcomes. In a similar vein, Tett and Burnett (2003) contend that personality traits are expressions of needs and are therefore intrinsically motivating, whereas abilities (skills) expressed in the workplace are motivated by extrinsic rewards. Moreover, the work of Klein, DeRouin, and Salas (2006) breaks down skill categories, such as communication, into more specific skills such as: active listening, oral communication, written communication, assertive communication, and nonverbal communication (as cited by Matteson et al., 2016, p. 81). Therefore, it is possible not only to differentiate soft skills from personality traits but also to break down the skill clusters (or categories) into smaller subsets. In attempting to measure tacit skills, however, it would be difficult to capture a deeper level of detail than the causal mapping method allows without observing employees in action and/or participating in company meetings. We could then potentially describe precisely how a brand manager, faced with a specific situation (e.g. simultaneous demands from angry clients

193

and limited manpower to deal with those demands), prioritizes under stress. However, the researcher may find that even close observation reveals little because the brand manager prioritizes under stress stoically, i.e. expressing little to his or her co-workers. At any rate, the causal mapping method pulls in the opposite direction. The aim is to extract from specific examples and stories something which is general enough to be communicated beyond the specific circumstances, and which can be considered causally linked to some larger phenomenon such as organizational success. While observation and participation have the potential to enrich research on tacit skills, and to capture levels of detail inaccessible to the researcher during semistructured interviews, the causal mapping method is coherent with the research aim of seeking a reasonable degree of generalization. Bearing in mind the inherent challenges and limitations reviewed above, and bearing in mind the intrinsic research aims of causal mapping, an effective assessment of the strengths and weaknesses of the method now can be meaningfully attempted. First, causal mapping surfaces tacit skills from multiple points of view in the organization without observation or participation; this can be considered a strength because it is less time consuming for the researcher and less intrusive for the company. The method allows the researcher to gather comments, observations, insights, and stories across multiple job functions, departments, and levels of hierarchy. This is coherent with the aim of capturing a reasonable degree of generalization because the researcher does not get exposed to the details of daily practice. Second, as an exercise in collective sense making, the individual and group interviews are a form of socialization as per Nonaka (1994), and a form of interpretation as per Crossan et al. (1999). This is clearly also a strength because the semi-structured interviews encourage knowledge sharing and organizational learning; they

194

provide insights to both the participants and the researcher about tacit skills and organizational success (Ambrosini & Bowman, 2001). Finally, an additional strength of the causal mapping method is the iterative process it allows for data gathering and analysis, as described already in the Chapter 3 section Design Overview. The researcher can continuously adjust his or her data gathering approach (i.e. interview questions). If the questions asked during one interview do not lead to useful data, the researcher can modify the questions in a subsequent interview. Thanks to this flexibility, the interview question strategy in this study was modified for Group Interview 5 after consultation with the research supervisor. This modification was motivated by the conclusion (discussed more below), that the responses to the interview questions in prior interviews sometimes included too many generalizations, espoused beliefs, and personal views or frustrations of individual employees to effectively reveal tacit knowledge. Hence a strategy was adopted during Group Interview 5 to circumvent these types of responses by asking participants more direct questions about their daily work routines and coping strategies when faced with problems. One of the weaknesses of the causal mapping method, as implied above, is that occasionally discussions get side tracked in ways that don’t yield useful information about tacit skills. Because of the exploratory nature of the method, it is hard for the researcher to know this is happening during the interviews themselves—it becomes apparent when reviewing the interview transcripts and analyzing the data. This weakness is very close to an inherent methodological challenge because a successful result depends largely on the researcher’s ability to navigate between two imperatives which are in apparent contradiction: the need for participants to feel comfortable expressing themselves freely and openly, on the one hand; and, the need for the researcher to guide and structure the discussion in specific ways, on the other.

195

For example, participants sometimes want to talk about what they perceive to be problems in the company. Talking about such problems may reveal useful information about the participant’s knowledge, practices and problem solving strategies; but, then again, it may only reveal the participant’s frustrations and personal views. In practice, it is hard to separate the two, and it is impossible for the interviewer to know in advance which way the discussion will go. Furthermore, companies have espoused values and beliefs that employees feel comfortable talking about, but which are not necessarily accurate reflections of the company culture or practices—in fact they may even mask the company’s true culture and practices (Schein, 2010). The researcher must therefore be careful not to accept everything people say during the interviews at face value. An additional reason why the interviews do not always yield useful information about tacit skills is because it is easier for managers and employees to speak about their explicit knowledge than it is to speak about something they are not used to speaking about, and/or something that is taken for granted because it is so embedded in practices and routines. Like the weakness described in the preceding paragraph, this one is also very close to a methodological challenge, and it depends greatly on the interviewer’s skill to mitigate its potential effects. Integrating the knowledge-in-practice framework (McIver et al., 2013) into the semi-structured interviews appears to have had both positive and negative impacts on the causal mapping method with respect to achieving a productive balance between explicit and tacit skills. While the framework provided an additional element of structure to guide the interview questions and the analysis of the results, it may also have oriented the discussion too much toward the job content and skills required for specific positions; thus, it may have inadvertently elicited explicit knowledge and skills rather than tacit ones. Indeed, the analysis using the framework revealed

196

fewer tacit skills than hoped for. An alternative explanation, however, for this somewhat disappointing result is that the company’s activity requires employees who are highly trained for specific technical tasks but not necessarily experienced. This lack of experience, as previously noted, is a characteristic of many young technology companies, and it compounds the inherent difficultly of getting people to reveal tacit knowledge and skills. To conclude, given a certain number of inevitable challenges to studying tacit skills and knowledge, the causal mapping method is well suited to assessing this difficult-to-define and difficult-to-quantify concept. Despite some methodological weaknesses noted above, the results of the semi-structured interviews provided valuable data for this mixed methods study. What is more, causal mapping could be utilized to study similarly difficult-to-quantify topics such as organizational culture, and teamwork, to the extent that the research goals are coherent with a reasonable degree of generalization. A deeper level of detail and a richer appreciation of the context would be possible through fieldwork; therefore, depending on the research objectives, the scope of the study, and the resources available, fieldwork would be an effective way to complement the semi-structured interview approach. Research. There are many opportunities for future research to advance our knowledge of organizational learning and knowledge management germane to the quest for competitive advantage, both in general, and specifically concerning the e-commerce sector. Areas which are ripe for future studies include company formalization and its impact on organizational learning, life cycle phases of development of e-commerce companies, communities of practice, as well as issues of complexity and imitability. Several of these are considered in turn below. The first subsection deals with organizational learning and knowledge management, in general, while the second deals specifically with the e-commerce sector.

197

Future research pertaining to organizational learning and knowledge management. In this chapter’s Theory subsection, the relationship between organizational learning, degrees of formalization and company structure, and life cycle phases of development were discussed. Because there appear to be few published studies which deal directly with this relationship, there are opportunities here for future research. For example, a promising research approach would involve comparing the learning environments and processes of two or more organizations operating in the same sector, but which exhibit very different characteristics pertaining to their respective stages of development—i.e. different sizes, ages, degrees of formalized structures, et cetera. Such studies might attempt to assess the extent to which formalization of company structure and processes is an antecedent for certain types of learning practices and processes. A multiple case study approach employing mixed methods would likely be effective for this. A related question dealing with life cycle phases of development, which was already raised in this chapter’s Theory subsection, is whether or not the learning life cycle in a young company can progress to concrete processes and practices, and to formalized knowledge management practices, without the company entering a formalization phase according to Daft’s (2010) model—i.e. without the company experiencing a crisis in management and leadership. A related question, also raised previously, is whether or not the company sacrifices potential growth and efficiency by not entering a formalization phase. Addressing these questions would require a longitudinal case study approach, ideally comparing two or more companies which had reasonably similar beginnings but made different choices and thus evolved in distinct ways. Another promising avenue for research would focus on the learning processes and practices associated with specific structures or knowledge management tools. For example, much has been written in the knowledge management literature about communities of practice,

198

which are considered to be “effective tools to capture and exchange both explicit and tacit knowledge” (Yamklin & Igel, 2012). Brown and Duguid (1991), referring to the work of Lave and Wenger (1991), contend that communities of practice are informal groups which form naturally and spontaneously outside the formalized company structure, and they argue that these informal groups are crucial for organizational learning and innovation—even more crucial, in fact, than formal company structures. Wenger and Snyder (2000) later outline a concept of communities of practice as a knowledge management tool which can be cultivated and managed with specific organizational goals in mind. This notion is taken further by Garavan, Carbery, and Murphy (2007) and by Yamklin and Igel (2012) who develop the concept of purposefully created communities of practice with structures and agendas defined by the company. Bearing in mind these different interpretations of the concept, future studies might involve close observation of communities of practice with a view to assessing the extent to which group learning is taking place using the criteria proposed by Wilson et al. (2007). Observation might also try to capture how communities of practice store and retrieve knowledge which they generate. Also, future studies could attempt to establish whether communities of practice are engaged in incremental or radical learning as per Edmondson’s definition (2002), and/or whether Edmondson’s (1999) theory of Psychological Safety applies in communities of practice in the same way she proposes it impacts work teams. Moreover, taking into consideration the work of Crossan et al. (1999), studies of communities of practice could aim to determine what level of learning processes are active in these groups, and the extent to which they facilitate the upward movement of tacit knowledge and innovation from individuals to the larger organization (feedforward), or the downward movement of innovation towards individual assimilation (feedback), or both. Finally, although there appear to be several studies measuring

199

success factors, and enabling factors for communities of practice (Jagasia et al., 2015; Yamklin & Igel, 2012; Yang & Wei, 2010), there are few studies which attempt to assess the impact of communities of practice on organizational learning outcomes. Future research pertaining to organizational learning and knowledge management in ecommerce companies. As stated in this chapter’s Theory subsection, there is little research available on life cycle development of e-commerce companies. Thus a promising direction for future research would be to explore this question. Studies might, for example, examine several e-commerce companies which have experienced high growth, and assess the extent to which they exhibit characteristics of Daft’s (2010) life cycle phases of development; and, furthermore evaluate the extent to which they have remained competitive in their respective markets. Likewise, there is little research to be found on communities of practice in e-commerce companies—most examples seem come from product development, and manufacturing (Yamklin & Igel, 2012) or IT solutions (Lee, Reinicke, Sarkar, & Andersen, 2015; Pandey & Dutta, 2013). Hence, studies assessing the presence, role, and significance of communities of practice in ecommerce companies would be a valuable contribution to the field. Another key question which future research might address is the issue of complexity and imitability in the e-commerce sector. According to Rivkin (2001), company tacit knowledge, personal relationships, and routines, are difficult to imitate. Furthermore, Rivkin defines productive systems as being complex if they consist of numerous elements, and if those elements interact with each other “richly” (p. 274); he conjectures that strategies of moderate complexity allow for replication but make imitation by rivals difficult. A natural extension of the research done for this dissertation would be to explore the extent to which knowledge in e-commerce companies is transferable without being imitable; and/or to assess the complexity of productive

200

systems in e-commerce companies. For example, studies could examine how managerial decisions are entwined with the technology, and/or how intertwined the specific client relationships are with the managerial decisions, and with the technology. Another potential research topic worthy of exploring is knowledge leaks. E-commerce has some unique features, described in the Chapter 2 subsection E-commerce differentiated, which suggest this issue may be particularly relevant. The technology evolves quickly in ecommerce but is relatively accessible and knowable to other organizations. In the study company, most employees are young, inexperienced, and there is a high rate of staff turnover; furthermore, there is considerable knowledge sharing which transpires between the company employees, and the employees of the client brands. Research which attempts to establish how much employee mobility and knowledge sharing are normal for the sector, and/or which examines the potential risks of knowledge leaks for companies would help advance our understanding of knowledge management and learning in e-commerce companies. Practice. This subsection begins by discussing suggestions to improve the functioning and effectiveness of the study company; the discussion should be considered in light of the cycles of knowledge creation and learning processes as presented in Figures 20-22. Suggestions for the company include creating communities of practice to encourage socialization and knowledge sharing; identifying areas where tacit knowledge of individuals needs to be explicitly shared and stored; identifying and spreading best management practices; developing a framework to provide more effective guidance around the concept of client autonomy; addressing counterproductive informal learning and work practices; investing in training programs that will help push certain types of organizational learning down from the organizational level to the individual level; and, finally, leveraging the unexploited knowledge

201

embedded in the after-sales service. The second part of this subsection discusses suggestions to improve the functioning of e-commerce companies in general. These suggestions include making investments in learning processes, training programs, and knowledge management tools as a means to developing a long-term sustainable competitive advantage; embedding organizational learning in the workplace through project-based learning; adopting a moderately complex strategy to reduce the risk of imitation while allowing for some replication; and developing these strategies in such a way that technology is intertwined with practices, tacit skills, and knowledge. Suggestions to improve the functioning and effectiveness of the study company. The study company is going through a crucial phase in its development. The capabilities of the company are migrating from resources such as individuals and individual tacit skills, towards more visible processes and routines; the culture and values are coalescing. According to Christensen (1997/2011), once this process is more advanced, it will be difficult to change. Therefore, the company leaders need to clearly assess where there is a good match between, on the one hand, the existing or emerging processes and routines, and, on the other, the set of problems the company is currently facing in its external environment. Furthermore, if the company has ambitions to become a hub of knowledge in the local e-commerce apparel sector, it must be as aware as possible of the knowledge it already possesses and has the potential to generate. As it weighs various measures to take, the cycles of knowledge creation and learning processes should be considered, as presented in Figures 20-22. The first recommendation to improve the functioning and effectiveness of the study company is for the enterprise to encourage the sharing of tacit knowledge (skills), and individual intuitions through socialization. While the semi-structured interview data suggest there is a

202

significant amount of employee-to-employee learning, much of this learning appears to be an informal and improvised strategy to face challenging circumstances such as high staff turnover and conflicting client demands. The company should thus create more explicit opportunities for socialization and group reflection such as communities of practice which are purposefully created and have specific goals (Garavan et al., 2007; Wenger & Snyder, 2000; Yamklin & Igel, 2012); this is particularly important for employees of the e-commerce department who the findings show to be under intense time pressure and thus lacking in opportunities for reflection. In addition to encouraging knowledge sharing, communities of practice might spur employees to identify more with the company’s culture and stated values which, in turn, should help ensure that individual innovations are in line with the company’s values and objectives (Crossan et al., 1999). Furthermore, the communities of practice may prove effective for problem solving (Wenger & Snyder, 2000) and helping employees develop strategies for facing uncertainty. Another recommendation for the company is to identify areas where tacit knowledge of individuals needs to be shared and stored in a more explicit format. An inventory of skills that are necessary for the basic functioning of the company, such as anything related to the fundamental technical creation and maintenance of the websites, should be drawn up; measures should be taken to ensure such skills do not reside solely in individuals who may at any time leave the company and take their knowledge with them. Such measure would likely involve some form of formal employee-to-employee transmission of knowledge possibly through mentor-apprentice type schemes and/or through documentation and storage on an intranet, or other appropriate and accessible resource. The company must also ensure that the stored knowledge is effectively retrieved, disseminated and applied so they lead to improvements in organizational performance over time (Wilson et al., 2007)

203

In a similar vein, best management practices should be identified, formally recognized and spread more evenly across the company. To take the example previously discussed of the brand manager who develops a sense of autonomy among her e-store team members by asking them the “right questions”, such behaviors which help to develop a stated value of the company, and to embed this value into practices and routines, can be disseminated to other brand managers as well as down from top managers to lower-level managers and employees. The results of this study also suggest more should be done to clarify and provide guidance around the concept of client development and autonomy, an organizational construct identified in the semi-structured interviews. The construct fits Schein’s (2010) definition of an espoused value—it is frequently spoken of, but the interview data suggests it is difficult to interpret and to apply in daily practice. Employees, particularly e-store managers, must face considerable diversity in their clients’ familiarity and confidence with e-commerce, as well as in their clients’ expectations about, and attitudes towards the study company. Tacit knowledge about specific client relationships are intertwined with decisions about services offered, whether those services involve expert advice, organizing sales and promotions, and/or delivering technical features of websites. Although the rubric, and the new approach to onboarding clients which has recently been implemented by the sales team, are meant to address some of the inherent ambiguity in the relationships with the brands, a framework should be developed to guide employees on a day-to-day basis, especially considering that many of the company’s current clients were brought on board before the rubric was implemented, and before the new approach was introduced. One possible way to address this would be to have e-store managers and brand managers conceptualize client clusters—i.e. groups of clients who share certain expectations and characteristics. These clusters might resemble a life cycle concept whereby

204

clients are grouped according to how experienced they are with e-commerce, and the extent to which they have developed a strategic understanding of e-commerce. Sets of guidelines for employees could then be developed and tailored to each client cluster. Also, brand managers and their e-store teams might be assigned specific clusters so that they become somewhat specialized in dealing with these client subsets. Developing a scheme like this would encourage the sharing of tacit knowledge and experience with a view to generating new, explicit knowledge to guide practice on a company-wide level. According to the learning organization literature (Argyris & Schon, 1996; Caldwell, 2012), learning is not always good. Christensen (1997/2011) warns that capabilities can become “disabilities” if they are not effectively aligned with the current set of problems the company is facing. For example, as already noted, the semi-structured interview data and the survey results suggest that e-commerce employees are under intense time pressures. In Group Interview 3, two junior e-store managers describe one of the ways they cope with the challenges of delivering services to demanding clients under tight deadlines. As discussed in the Chapter 4 section Answers to Research Questions, these managers use the term “free style” to denote how they choose to skip certain standard procedures and processes in order to deliver websites and website features on time; however, these sites are delivered with known bugs which have to be resolved after the sites go live. The recurrent nature of related concerns in the interview data suggest that the term “free style” captures an underlying assumption among many employees that they have to cut corners to meet deadlines to survive in their positions. When this issue was discussed with the e-commerce manager during the follow up interview, he stated that one of the major challenges he faced as a manager was the company’s inability to produce high quality websites under tight deadlines. Based on his past experience at another company which was able to do

205

this, Abel believes this is an achievable goal. According to him, tight deadlines are not the problem but rather lack of team coordination, communication, and stable team membership. Hence, it is not entirely clear whether the best approach for addressing this issue would focus on improved team coordination, or better managing the workload to alleviate the time pressures—or if the two approaches could be reconciled somehow. At any rate, the current situation involves informal learning that will be unproductive in the long run, and should thus be corrected before it becomes a “disability”. Some of the difficulties which surfaced in the semi-structured interviews require organizational learning on the institutional level. Lack of communication across departments and teams, workflow and capacity problems, for example, require a structural approach which is likely to come from the top down. It is difficult for an outsider to make specific recommendations here but a few general ideas are evoked. For example, project managers should be systematically consulted before deals are signed to onboard new clients and/or projects—this is now done occasionally, according to the e-commerce manager, but it is neither formal nor systematic. Also, clear systems and chains of communication need to be designed to inform clients when there are technical problems which will impact the performance of their websites. More generally, the company should invest in training programs and initiatives that will push learning processes from the organizational level down to the individual level. Such investments will help address the tension which exists in the company between short-term commercial goals and longer-term strategic innovation goals in favor of the latter. Training programs might emphasize the importance of storing and sharing knowledge or might provide employees with tools to face highly uncertain and stressful situations. They might also emphasize the value of transversal communication and involve some form of staff rotation, as

206

suggested above. Similar to socialization tools, such as communities of practice, training programs have the potential to cultivate a stronger identification between employees and the company, and to facilitate the assimilation of values (Crossan et al., 1999). This, in turn, is likely to reduce staff turnover, which has clearly been a problem in the company. In addition, metaphors, such as looking under the hood of the car, may be employed by the company’s leadership to communicate and spread hard-to-define tacit skills like thinking holistically and proactively in order to solve problems (Ambrosini & Bowman, 2001; Nonaka, 1994). The last suggestion to improve the functioning and effectiveness of the study company attempts to address a striking example of knowledge absence in the study company. The management does not appear to take any serious interest in the potential wealth of knowledge embedded in the after sales service activity. This service seems to be the most isolated of all the departments in the company—it is visually distinct from the rest of the departments in the way the employees sit in one defined area at their mini-cubicles wearing headphones and talking to customers. Although the manager of this service participated in an individual interview, no employees from this service participated in the group interviews, and there is no evidence that after sales service employees participate in the cross functional work teams or interact in any way with the other employees. When questioned about this during the follow up interview, the e-commerce manager, explained that e-store managers had previously been tasked with the responsibility of answering calls and electronic queries from customers, but when the volume became too high, this activity was split off into a separate service. Today, the after sales service handles hundreds of queries each day from the final consumers of the apparel brands, so there is indeed no realistic way e-store managers could handle the work load. However, the result of this evolution is that the two branches of customer service—one for the client brands, one for the

207

final customers—are now completely separate, with the latter effectively isolated from the rest of the company’s activities. The considerable knowledge embedded in the after sales service activity about consumer’s preferences, attitudes and behaviors is left unexploited. This knowledge could complement the regular quantitative analyses the study company runs on web traffic and online purchases. Additionally, an analysis of the after sales service data could be leveraged as an additional service to be offered to the brands. With first-hand access to data about the consumer preferences of 30 different apparel brands, the study company is in a strong position to do comparative analyses that would be valuable to their clients. Hence seeking explicit ways to exploit this potential knowledge, and better integrating employees from the after sales service into work teams, and possibly even project teams, would like be beneficial for the learning and knowledge management of the company. The company’s management might also consider rotating employees from other departments to do shifts in after sales service to gain direct contact with the final consumers and to better mitigate this functional separation of final consumer services from client brand services. Suggestions to improve functioning in e-commerce companies in general. E-commerce companies should not assume that competitive advantage necessarily lies in the explicit technical knowledge and skills of the company and of its employees. Such knowledge is likely to be transferable, which means easy to imitate and easy to lose when employees who possess such skills and knowledge leave the company. While explicit technical knowledge may be the key to gaining advantages early in the company’s history, especially if the enterprise is one of the early movers in a given market, the longer-term approach to knowledge and competitive advantage requires focusing on how people work together, on the specific relationships the company has with clients and/or suppliers, and on the company’s ability to learn from the external

208

environment in order to evolve as circumstances change. As mentioned in the Chapter 2 subsection Knowledge management, Spender (1993) posits that even the development and patenting of cutting-edge breakthrough technology does not necessarily translate into business success. Therefore, investments in building learning processes and practices, training programs, and knowledge management tools may be key to the company’s long-term survival and success. In order to reduce the risk of an e-commerce company’s strategy getting copied, companies in the sector should aim for a moderate degree of complexity (Rivkin, 2001). Many of the platforms, systems, and technical tools used in e-commerce are shared by other actors in the sector, and the skills required to use them can be easily learned. Although in France, ecommerce companies may struggle to find properly trained employees because currently the demand is higher than supply, this situation is likely to be temporary; furthermore, it only reinforces the need to invest in learning processes and training programs which help reduce staff turnover, and subsequently reduce knowledge leaks. The example of the study company suggests that an enterprise which is successful in offering e-commerce services to a client brand can replicate the strategy in order to significantly expand its client portfolio while staying in a given market and sector; the head of commercial development and strategy estimates the study company has no real competitors in the apparel sector in its national market, and argues that the company’s added value is its technical and marketing expertise (see Individual interview 3). This appears to fit Rivkin’s (2001) definition of moderate complexity: the strategy is not so complicated that it renders effective replication impossible, but it is complicated enough that competitors have not been successful at imitating it. The combination of technical and marketing expertise probably explains this moderate degree of complexity. This expertise lies in the company’s deep understanding of the rich and numerous

209

interactions required between consumers, the brand, and the technology. The extent to which this model can be replicated on a wider scale internationally has yet to be tested but would likely require facing some daunting challenges due to need for on-the-ground knowledge and experience with apparel brands, tastes, and supply chain logistics in other localities. As stated in the Chapter 2 subsection Competitive advantage in e-commerce enterprises, people, processes, and practices cannot be dissociated from the technology, nor can their role be diminished. E-commerce companies must develop strategies which deal with technology as intertwined with practice, tacit skills and knowledge. Furthermore, they must develop strategies which evolve hand-in-hand with employees’ knowledge of the market, and the rich interactions among the different players. Although these principles should apply to all aspects of ecommerce, when companies manage the B2C e-commerce for B2B clients—as is the case with the study company—the importance of the specific relational knowledge, and the knowledge stemming from the rich interactions among the different players will be even greater compared to companies that deal exclusively with either B2B e-commerce or B2C e-commerce, Lastly, e-commerce companies would do well to heed the recommendations of Gunasekara (2003) and Raelin (1997) to embed organizational learning in the workplace. Gunasekara proposes a project-based workplace learning model which appears relevant to the ecommerce sector; he calls for establishing learning goals as critical outcomes of project management beyond the traditional outcomes which tend to focus only on efficiency and/or client satisfaction. Projects can be seen as opportunities for the organization to develop its core capabilities; this, of course has implications for the role of the project manager and the skills required to succeed in this role. Gunasekara argues that project management increasingly requires skills which are close to many of the tacit skills identified in this study, including the

210

ability to think holistically to solve problems (systems thinking), and the ability to deal with uncertainty and ambiguity. The more recent work of Lee et al. (2015) provides evidence that having project managers participate in communities of practice is not only beneficial to the individual project managers but is also improves knowledge sharing on an organizational level. Concluding Remarks Limitations of this study. As described in Chapter 1, there are a number of limits to this study. Because the research focused on one company, the applicability of the findings will be limited especially where the industry, employees, and culture are different. Also, the study made use of a nonrandom sampling technique which may hinder the generality of the results and conclusions; and sample sizes did not fully allow for statistical test of differences. The single cross-section descriptive survey and the semi-structured interviews transpired over a period of approximately two months. The data was therefore gathered in a fixed period at a company that is constantly changing, and which operates in a sector experiencing rapid and continuous transformation, as described in the Chapter 2 section Introduction to the E-commerce Industry. Furthermore, the quantitative and qualitative data were integrated by a single researcher increasing the likelihood of observer bias. Finally, as discussed in Chapter 2: Literature Review, there are inherent challenges to operationalizing and conducting reliable and repeatable empirical research on the concept of tacit knowledge, not to mention on the constructs of organizational learning and knowledge creation; there are also theoretical weaknesses in the concept of the learning organization. Given the lack of broad agreement about the nature of knowledge and how to study it, the methodology and findings of this study may not be acceptable to all. Expression of appreciation to the study company. This study set out to explore the links between tacit knowledge, organizational learning, and competitive advantage in an e-

211

learning company using a mixed methods case study approach. The research was facilitated by the leadership of the study company particularly by the e-commerce manager, and the human resources manager both of whom gave backing to the study, participated in the individual interviews, helped to organize the group interviews with the employees, and participated in a follow up interview where findings were discussed. I would therefore like to express my appreciation to them and to the study company for their help and cooperation.

212

References 2 billion consumers worldwide to get smart (phones) by 2016. (2014, December 11). eMarketer. Retrieved from http://www.emarketer.com/Articles/Print.aspx?R=1011694 Aggestam, L. (2006). Learning organization or knowledge management - Which came first, the chicken or the egg? Information Technology and Control, 35(3A), 295-302. Alderfer, C. P. (1987). An intergroup perspective on organizational behavior. In J. W. Lorsch (Ed.), Handbook of Organizational Behavior (pp. 190-222). Englewood Cliffs, NJ: Prentice-Hall. Ambrosini, V., & Bowman, C. (2001). Tacit knowledge: Some suggestions for operationalization. Journal of Management Studies, 38(6), 811-829. An, Y.-J., & Reigeluth, C. M. (2005). A study of organizational learning at smalltown hospital. Performance Improvement, 44(10), 34-39. Andreu, R., & Ciborra, C. U. (1996). Core capabilities and information technology: An organizational learning approach. In B. Moingeon, & Edmonson, A. (Eds.), Organizational Learning and Competitive Advantage. New York, NY: Sage. Argote, L. (2005). Reflections on two views of managing learning and knowledge in organizations. Journal of Management Inquiry, 14(1), 43-48. Argote, L. (2011). Organizational learning research: Past, present and future. Management Learning, 42(4), 439-446. Argote, L., & Ingram, P. (2000). Knowledge transfer: A basis for competitive advantage in firms. Organizational Behavior and Human Decision Processes, 82(1), 150-169. Argote, L., & Miron-Spektor, E. (2009). Organizational learning: from experience to knowledge. New Perspectives in Organization Science. Pittsburgh, PA: Carnegie Mellon University. Argote, L., Gruenfeld, D. H., & Naquin, C. (2001). Group learning in organizations. In M. E. Turner (Ed.), Groups at Work: Advances in Theory and Research (pp. 369-411). Mahwah, NJ: Lawrence Erlbaum Associates. Argyris, C. (2002). Double-loop learning, teaching, and research. Academy of Management Learning & Education, 1(2), 206-218. Argyris, C., & Schon, D. A. (1996). Organizational Learning II: Theory, Method, and Practice, Volume 2. University of Michigan, MI: Addison-Wesley Publishing Company. Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-210.

213

Barney, J., Wright, M., & David J. Ketchen, J. (2001). The resource-based view of the firm: Ten years after 1991. Journal of Management, 27(6), 625. Beattie, A. (n.d.). Market crashes: the dotcom crash. Retrieved December 27, 2016 from http://www.investopedia.com/features/crashes/crashes8.asp?ad=dirN&qo=investopediaSi teSearch&qsrc=0&o=40186 Berger, P. L., & Luckmann, T. (1966). The Social Construction of Reality. Garden City, NY: Doubleday. Bettis, R. A., & Hitt, M. A. (1995). The new competitive landscape. Strategic Management Journal, 16, 7-19. Bhat, J. S., & Kumar, V. (2011). A structured approach to knowledge management in SMEs: towards a successful manufacturing strategy. International Journal of Business Performance Management, 6(3/4), 1. Bheemaiah, K. (2015, January ). Block Chain 2.0: The renaissance of money . Retrieved January 1, 2017 from https://www.wired.com/insights/2015/01/block-chain-2-0/ Brown, J. S., & Duguid, P. (1991, February). Organizational learning and communities-ofpractice: Toward a unified view of working, learning, and innovating. Organizational Science, 2(1), 40-57. Brown, J. S., & Duguid, P. (2001). Knowledge and organization: A social perspective. Organization Science, 12, 198-213. Brown, M. (October 27, 2015). B2B eCommerce: the trillion dollar industry. Allen, TX: PFSweb. Brynjolfsson, E., & Kahin, B. (2000). Introduction. In E. Brynjolfsson, & Kahin, B. (Eds.), Understanding the Digital Economy: Data, Tools, and Research (pp. pp. 1-10). Cambridge, MA: MIT Press. Bui, H., & Baruch, Y. (2010). Creating learning organizations: A systems perspective. The Learning Organization, 17(3), 208-227. Caldwell, R. (2012, June). Systems thinking, organizational change and agency: A practice theory critique of Senge's Learning Organization. Journal of Change Management, 12(2), 145-164. Christensen, C. M. (2011). The innovator's dilemma: The revolutionary book that will change the way you do business. New York, MA: Harper Business Essentials. (Originally published in 1997). Christensen, L. B., Johnson, R. B., & Turner, L. A. (2011). Research Methods, Design, and Analysis. Boston, MA: Pearson Education Inc.

214

Ciborra, C. U., & Andreu, R. (2001). Sharing knowledge across boundaries. Journal of Information Technology, 16(2), 73-81. Company Info. (n.d.). Retrieved January 1 2017, 2017 from https://newsroom.fb.com/companyinfo/ CompuServe unveils the 'electronic mall'. (1984, September 11). Marketing News, 18(23), 3. Cook, S. D., & Brown, J. S. (1999). Bridging epistemologies: The generative dance between organizational knowledge and organizational knowing. Organization Science, 10(4), 381400. Creswell, J. W. (2011). Designing and Conducting Mixed Methods Research (2nd ed.). Thousand Oaks, CA: Sage Publications Inc. Cronbach, L. J. (1989). Construct validation after thirty years. In R. L. Linn (Ed.), Intelligence: Measurement, Theory, and Public Policy (pp. 147-171). Urbana and Chicago, IL: University of Illinois Press. Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Pscyhological Bulletin, 52, 281-302. Crossan, M. M., Lane, H. W., & White, R. E. (1999). An organizational learning framework: From intuition to institution. Academic of Management Review, 24(3), 522-537. Crotty, M. (1998). The Foundations of Social Research. London: Sage. Daft, R. L. (2010). Organization Theory and Design (10th ed). Mason, OH: South-Western Cengage Learning. DHgate.com. (n.d.). Retrieved December 27, 2016 from http://research.omicsgroup.org/index.php/DHgate.com Dodgson, M. (1993). Organizational learning: A review of some literatures. Organizational Studies, 14(3), 375-394. E-commerce statistics. (2016, December ). Retrieved January 1, 2017 from http://ec.europa.eu/eurostat/statistics-explained/index.php/E-commerce_statistics Eden, C., & Ackermann, F. (1998). Making Strategy. London: Sage. Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Sciences Quarterly, 44, 350-383. Edmondson, A. C. (2002). The local and variegated nature of learning in organizations: A grouplevel perspective. Organization Science, 13(2), 128-146.

215

Edmondson, A. C. (2008, July-August). The competitive imperative of learning. Harvard Business Review. Retrieved from https://hbr.org/2008/07/the-competitive-imperative-oflearning Ellis, A. P., Hollenbeck, J. R., Ilgen, D. R., Porter, C. O., & West, B. J. (2001). Team learning: Collectively connecting the dots. Journal of Applied Psychology, 88, 821-832. Evangelista, F., Poon, P., & Albaum, G. (2012). Using response behaviour theory to solicit survey participation in consumer research: An empirical study. Journal of Marketing Management, 28(8-10), 1174-1189. Evans, P., & Wurster, T. S. (1999, November December). Getting real about virtual commerce. Harvard Business Review. Retrieved from https://hbr.org/1999/11/getting-real-aboutvirtual-commerce Fiol, C. M., & Lyles, M. (1985). Organizational learning. Academy of Management Review, 10(4), 803-813. Forrester, J. W. (2007). System dynamics—A personal view of the first fifty years. System Dynamics Review, 23(2/3), 345-358. Foss, N., & Mahnke, V. (2003). Knowledge management: What can economics management contribute. In M. Easterby-Smith, & Lyles, M. (Eds.), The Blackwell Companion to Organizational Learning and Knowledge Management (pp. 78-103). Oxford: Blackwell. Garavan, T. N., Carbery, R., & Murphy, E. (2007). Managing intentionally created communities of practice for knowledge sourcing across organizational boundaries: Insights on the role of the CoP manager. The Learning Organization, 14(1), 34-49. doi:org/10.1108/09696470710718339 Garcia, T. (2015, December 22). Amazon will account for more than half of 2015 e-commerce growth, says Macquarie. Retrieved January 1, 2017 from http://www.marketwatch.com/story/amazon-will-account-for-more-than-half-of-2015-ecommerce-growth-says-macquarie-2015-12-22 Garratt, B. (1995). An old idea that has come of age. People Management, 1(19), 25. Garvin, D. A. (2000, July-August). Building a learning organization. Harvard Business Review. (Original work published in 1993). Retrieved from https://hbr.org/1993/07/building-alearning-organization Garvin, D. A., Edmondson, A. C., & Gino, F. (2008, March). Is yours a learning organization? Harvard Business Review. Retrieved from https://hbr.org/2008/03/is-yours-a-learningorganization Gendall, P. (1999). Responding to the problem of non-response. MRSA Conference, Adelaide.

216

Gereffi, G. (2001). Beyond the producer-driven/buyer-driven dichotomy: The evolution of global value chains in the internet era. IDS Bulletin, 23(3). doi:10.1111/j.17595436.2001.mp32003004.x Gibson, C., & Vermeulen, F. (2003). A healthy divide: Subgroups as a stimulus for team learning behavior. Administrative Science Quarterly, 48, 202-224. Goh, S. C. (1998). Toward a learning organization: The strategic building blocks. Sam Advanced Management Journal, 63(2), 15-22. Grant, R. (2013, August). Reflections on knowledge-based approaches to the organization of production. Journal of Management & Governance, 17(3), 541-558. Grant, R. M. (2010). Contemporary strategy analysis (7th ed). West Sussex, UK: John Wiley & Sons, Ltd. Greiner, L. E. (1972, July-August). Evolution and revolution as organizations grow. Harvard Business Review, 37-36. Retrieved from https://hbr.org/1998/05/evolution-andrevolution-as-organizations-grow Grimm, K. J., & Widaman, K. F. (2012). Construct validity. In H. Cooper, Camic, P. M., Long, D. L., Panter, A. T., Rindskopf, D., & Sher, K. J. (Eds.), APA Handbook of Research Methods in Psychology, Vol 1: Foundations, Planning, Measures, and Psychometrics (pp. 621-642). Washington, DC: American Psychological Association. Gruenfeld, D. H., Martorana, P. V., & Fan, E. T. (2000). What do groups learn from their worldliest members? Direct and indirect influence in dynamic teams. Organizational Behavior and Human Decision Processes, 82, 45-59. Guba, E. G. (1981). Criteria for assessing the trustworthiness of naturalistic inquiries. Educational Communication and Technology Journal , 29, 75-91. Gunasekara, C. (2003, Winter). Project-based workplace learning: A case study. SAM Advanced Management Journal, 68(1), 37-48. Hackman, R. J. (1987). The design of work teams. In J. Lorsch (Ed.), Handbook of Organizational Behavior (pp. 315-342). Upper Saddle River, NJ: Prentice-Hall. Hardy, Q. (2016, September 18). Artificial intelligence software is booming. But why now? Retrieved from http://www.nytimes.com/2016/09/19/technology/artificial-intelligencesoftware-is-booming-but-why-now.html?_r=0 Hegedus, N. (2009). Riding the waves: And understanding what hit us. Ericsson Business Review, 3, 10-15. Heorhiadi, A., La Venture, K., & Conbere, J. P. (2014). What do organizations need to learn to become a learning organization? OD Practitioner, 46(2), 5-9.

217

Hitt, M., Ireland, R., Sirmon, D., & Trahms, C. (2011, May ). Strategic entrepreneurship: Creating value for individuals, organization, and society. Academy of Management Perspectives, 25(2), 57-75. Hoffman, D. L., & Novak, T. P. (2000a, May June). How to acquire customers on the web. Harvard Business Review. Retrieved from https://hbr.org/2000/05/how-to-acquirecustomers-on-the-web Hoffman, D. L., & Novak, T. P. (2000b). The growing digital divide: Implications for an open research agenda. In E. Brynjolfsson, & Kahin, B. (Eds.), Understanding the Digital Economy: Data, Tools, and Research (pp. 245-260). Cambridge, MA: MIT Press. Hofstede, G. (2001). Culture's Consequences. Thousand Oaks, CA: Sage. Howe, N., & Strauss, W. (1991). Generations: The Hisotry of America's Future, 1584 to 2069. New York, NY: William & Morrow. Huber, G. P. (1991). Organizational learning: The contributing processes and literatures. Organization Sciences, 2(1), 88-115. https://doi.org/10.1287/orsc.2.1.88 Iankelevich, D. (2015). Worldwide retail ecommerce sales: Emarketer's updated estimates and forecast through 2019. New York, NY: eMarketer. Jagasia, J., Baul, U., & Mallik, D. (2015). A framework for communities of practice in learning organizations. Business Perspectives and Research, 3(1), 1-20. Jensen, E. (2005). A contextual theory of learning and the learning organization. Knowledge and Process Management, 12(1), 53-64. Johnson-Laird, P. N. (1996). Mental models, deductive reasoning, and the brain. In M. S. Gazzaniga (Ed.), The Cognitive Neurosciences (pp. 999-1008). Cambridge, MA: MIT Press. Kelly, K. (2005, August 1). We are the web. Retrieved from http://www.wired.com/2005/08/tech/?pg=2, Ed.) Wired. Kelly, K. (2016, May ). The untold story of magic leap, the world's most secretive start up. Retrieved from https://www.wired.com/2016/04/magic-leap-vr/ King, W. R. (2001, Winter). Strategies for creating a learning organization. Information Systems Management, 18(1), 12-20. https://doi.org/10.1201/1078/43194.18.1.20010101/31261.3 Kirwan, C. (2013). Making Sense of Organizational Learning. New York, NY: Routledge. Klein, C., DeRouin, R., & Salas, E. (2006). Uncovering workplace interpersonal skills: A review, framework, and reseach agenda. In G. P. Hodgkinson, & Ford, J. K (Eds.), International Review of Industrial and Organizational Psychology (pp. 79-126). West Sussex, UK: Wiley.

218

Klein, K. J., Dansereau, F., & Hall, R. J. (1994). Level issues in theory development, data collection, and analysis. Academy of Management Review, 19(2), 195-229. Krogh, G. V., Nonaka, I., & Aben, M. (2001). Making the most of your company's knowledge: A strategic framework. Long Range Planning, 34(4), 421-439. La Fédération du E-commerce et de la Vente à Distance. (2016). Bilan 2015 du e-commerce en France: les Français ont dépensé 65 milliards d'euro sur internet. Paris, France. Laudon, K. C., & Traver, C. G. (2017). E-commerce 2016: Business, technology, society (12th ed.). Essex, UK: Pearson Education Limited. Laumeister, G. (2014, August 20). Why online marketplaces are booming. Retrieved from http://www.forbes.com/sites/groupthink/2014/08/20/why-online-marketplaces-arebooming/print/ Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. New York, NY: Cambridge University Press. Lawrence, T. B., Mauws, M. K., Dyck, B., & Kleysen, R. F. (2005). The politics of organizational learning: Integrating power into the 4I framework. Academy of Management Review, 30(1), 180-191. Lazar, J. B., & Robu, D. (2015). Accelerating the development of learning organizations: Shifting paradigms. Theoretical and Applied Economics, 22(2), 241-256. Lee, L., Reinicke, B., Sarkar, R., & Andersen, R. (2015, February/March). Learning through interactions: Improving project management through communities of practice. Project Management Journal, 46(1), 40-52. doi:10.1002/pmj.21473 Levin, J. D. (2011, March). The economics of internet markets. Working Paper No. 16852, Cambridge, MA: National Bureau of Economic Research. London, M., Polzer, J. T., & Omoregie, H. (2005). Interpersonal congruence, transactive memory, and feedback processes. Human Resources Development Review, 4(2), 114-135. Marquardt, M. J. (1996). Building the Learning Organization. New York, NY: McGraw-Hill. Matteson, M. L., Anderson, L., & Boyden, C. (2016). "Soft Skills": A phrase in search of meaning. Libraries and the Academy, 16(1), 71-88. Matthews, P. (1999). Workplace learning: Developing an holistic model. The Learning Organization, 6(1), 18-29. McIver, D., Lengnick-Hall, C. A., Lengnick-Hall, M. L., & Ramachandran, I. (2012). Integrating knowledge and knowing: A framework for understanding knowledge-in-practice. Human Resource Management Review, 22(2), 86-99. https://doi.org/10.1016/j.hrmr.2011.11.003

219

McIver, D., Lengnick-Hall, C. A., Lengnick-Hall, M. L., & Ramachandran, I. (2013). Understanding work and knowledge management from a knowledge-in-practice perspective. Academy of Management Review, 38(4), 597-620. Médiamétrie. (2016, May 19). Baromètre trimestriel de l'audience du e-commerce en France [Press Release]. Retrieved from http://www.mediametrie.fr/internet/communiques/barometre-trimestriel-de-l-audiencedu-e-commerce-en-france-2eme-trimestre-2016 Merrill, J. B. (2016, August 23). Liberal, moderate or conservative? See how Facebook labels you. The New York Times. Retrieved from http://www.nytimes.com/2016/08/24/us/politics/facebook-adspolitics.html?hp&action=click&pgtype=Homepage&clickSource=storyheading&module=first-column-region®ion=top-news&WT.nav=top-news&_r=1. Messick, S. (1994, September). Validity of psychological assessment: Validation of inferences from person's responses and perofrmances as scientific inquiry into score meaning. ETS Report Series, (2). doi:10.1002/j.2333-8504.1994.tb01618.x Morse, J. M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2002, Spring ). Verification strategies for establishing reliability and validity in qualitative research. International Journal of Qualitative Methods, 1(2), 1-19. Nonaka, I. (1991/2007, July-August). The knowledge-creating company. Harvard Business Review, 85(7/8), 162-171. Nonaka, I. (1994, February). A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 14-37. Nusca, A. (2015, July 14). Pinterest CEO Ben Silbermann: We’re not a social network. Retrieved January 1, 2017 from http://fortune.com/2015/07/13/pinterest-ceo-bensilbermann/ Orlikowski, W. (2002). Knowing in practice: Enacting a collective capability in distributed organizing. Organization Science, 13(3), 249-273. https://doi.org/10.1287/orsc.13.3.249.2776 Orlikowski, W. J., & Iacono, C. S. (2000). The truth is not out there: An enacted view of the "digital economy". In E. Brynjolfsson, & Kahin, B. (Eds.), Understanding the Digital Economy: Data, Tools, and Research (pp. 352-380). Cambridge, MA: MIT Press. Pandey, S., & Dutta, A. (2013). Communities of practice and organizational learning: Case study of a globa IT solutions company. Strategic HR Review, 12(5), 255-261. https://doi.org/10.1108/SHR-02-2013-0010 Pedler, M., Burgoyne, J., & Boydell, T. (1991). The Learning Company: A Strategy for Sustainable Development. New York, NY: McGraw-Hill.

220

Perez, C. (2002 ). Technological Revolutions and Financial Capital: The Dynamics of Bubbles and Golden Ages. Cheltenham, UK: Elgar. Perez, C. (2009a, January). Technological revolutions and techno-economic paradigms. (T. U. Sussex, Ed.) Working papers in technology, governance, and economic dynamic, 20. Perez, C. (2009b). The double bubble at the turn of the century: Technological roots and structural implications. Cambridge Journal of Economics, 33, 779-805. doi:10.1093/cje/bep028 Peters, T. J., & Waterman, Jr., R. H. (1982). In Search of Excellence: Lessons from America's Best-Run Companies. New York, NY: Harper & Row. Petropolous, G. (2016, February 22). Uber and the economic impact of sharing economy platforms [Blog post]. Retrieved from http://bruegel.org/2016/02/uber-and-the-economicimpact-of-sharing-economy-platforms/ Pinsonneault, A., & Kraemer, K. L. (1993). Survey research methodology in management information systems: An assessment. Journal of Management Information Systems, 10(2), 75-105. Polanyi, M. (1966). The Tacit Dimension. London: Routledge. Prahalad, C., & Hamel, G. (1990, May-June). The core competence of the corporation. Harvard Business Review, 68(3), 79-91. Prusak, L. (2001). Where did knowledge management come from. IBM Systems Journal, 40(4), 1002. Quinn, R. E., & Cameron, K. (1983, January). Organizational life cycles and shifting criteria of effectiveness: Some preliminary evidence. Management Science, 29, 33-51. https://doi.org/10.1287/mnsc.29.1.33 Raelin, J. A. (1997, November-December). A model of work-based learning. Organization Science, 8(6), 563-578. Rayport, J. R., & Jaworski, B. J. (2000). E-Commerce. New York, NY: McGraw Hill/Irwin. Rivkin, J. W. (2001, May-June). Reproducing knowledge: Replication without imitation at moderate complexity. Organization Science, 12(3), 272-293. Rowden, R. A. (2001, Summer). The learning organization and strategic change. SAM Advanced Management Journal, 66(3), 11-16. Schein, E. (2004). DEC is Dead, Long Live DEC: The Lasting Legacy of Digitial Equipment Corporation. San Francisco, CA: Berrett-Koehler.

221

Schein, E. H. (2010). Organizational Culture and Leadership (4th ed.). San Francisco, CA: Jossey-Bass. Senge, P. (2006). The Fifth Discipline: The Art and Practice of the Learning Organization. New York, NY: Doubleday. (Orignal work published in 1990). Shapiro, C., & Varian, H. R. (1999). Information Rules: A Strategic Guide to the Network Economy. Boston, MA: Harvard Business School Press. Shaw, M., Bednall, D., & Hall, J. (2002). A proposal for a comprehensive response-rate measure (CRRM) for survey research. Journal of Marketing Management, 18(5-6), 533-554. https://doi.org/10.1362/0267257022683631 Shenton, A. K. (2004). Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22(2), 63-75. Shipton, H., Zhou, Q., & Mooi, E. (2013). Is there a global model of learning organizations? An empirical, cross-national study. The International Journal of Human Resources Management, 24(12), 2278-2298. Smits, S. J., & Bowden, D. E. (2015). A perspective on leading and managing organizational change. Economics and Business Review, 1(15), 3-21. Spender, J. (1993). Competitive advantage from tacit knowledge? Unpacking the concept and its strategic implications. Academy of Management Best Papers Proceedings, 37-41. Spender, J. (2014). KM: Origins, history and development. Advances in Knowledge Management, 1-20. doi:10.1007/978-3-319-09501-1_1 Stake, R. E. (2005). Qualitative case studies. In N. K. Denzin, & Lincoln, Y. S. (Eds.), The Sage Handbook of Qualitative Research (third ed., pp. 443-466). Thousand Oaks, CA: Sage. Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of job performance. Journal of Applied Psychology, 88(3), 500-517. Toffler, A. (1980). The Third Wave. New York, NY: William Morrow. Tozzi, J. (2008, July 11). Determining where to sell online. Retrieved from https://www.bloomberg.com/news/articles/2008-11-07/determining-where-to-sellonlinebusinessweek-business-news-stock-market-and-financial-advice US Census Bureau. (2016). E-stats 2014: Measuring the electronic economy. United States Census Bureau, Report No. E14-ESTATS. Retrieved from https://www.census.gov/library/publications/2016/econ/e14-estats.html Varian, H. (2000, February 21). 5 habits of highly effective revolutions. Forbes. Retrieved from http://www.forbes.com/asap/2000/0221/073.html

222

Wahba, P. (2015, July 20). Amazon to surpass Macy's as no.1 U.S. clothing retailer by 2017, analyst says. Retrieved from http://fortune.com/2015/07/20/amazon-macys-clothing/ Watkins, K. E., & Marsick, V. J. (1993). Sculpting the Learning Organization: Lessons in the Art of Systemic Change. San Francisco, CA: Jossey-Bass. Webster, C. (1997). Effects of researcher presence and appeal on response quality in handdelivered, self-administered surveys. Journal of Business Research, 38(2), 105-114. Weick, K. E. (1995). Sensemaking in Organizations. Thousand Oaks, CA: Sage. Weick, K. E., & Bougon, M. G. (1986). Organizations as cognitive maps: Charting ways to success and failure. In H. P. Sims, & Gioia, D. A. (Eds.), The Thinking Organization (pp. 102-135). San Francisco, CA: Jossey-Bass Inc. Wenger, E. C., & Snyder, W. M. (2000, January-February). Communities of practice: The organizational frontier. Harvard Business Review, pp. 139-145. Wilkens, A. L., & Thompson, M. P. (1991). On getting the story crooked and straight. Journal of Organizational Change Management, 4(3), 18-26. Williams, B. (2015, Janaury 29). What does the future hold for Bitcoin? [Blog post]. Retrieved from http://insights.wired.com/profiles/blogs/what-does-the-future-hold-forbitcoin#axzz4Ky8xPH10 Wilson, J. M., Goodman, P. S., & Cronin, M. A. (2007). Group learning. Academcy of Management Review, 32(4), 1041-1059. Winterman, D., & Kelly, J. (2013, September 16). Online shopping: The pensioner who pioneered a home shopping revolution. BBC News Magazine. Retrieved from http://www.bbc.com/news/magazine-24091393 Yamklin, S., & Igel, B. (2012). Communities of practice purposefully designed for improving business performance. Knowledge and Process Management, 19(4), 189-202. Yang, B., Watkins, K. E., & Marsick, V. J. (2004). The construct of the learning organization: Dimensions, measurement, and validation. Human Resource Development Quarterly, 15(1), 31-55. Yang, C.-L., & Wei, S.-T. (2010, October ). Modelling the performance of CoP in knowledge management. Total Quality Management, 21(10), 1033-1044. Yin, R. K. (2013). Case Study Research: Design and Methods (Applied Social Research Methods). Thousand Oaks, CA: Sage. YStats.com. (2016). Europe B2C e-commerce market 2016. Retrieved from https://www.ystats.com/infographic-western-europe-b2c-e-commerce-market-2016/

223

Zander, U., & Kogut, B. (1995). Knowledge and the speed of the transfer and imitation of organizational capabilities: An empirical test. Organization Science, 6(1), 76-92. Zaroban, S. (2015, January 13). Social networks deliever for retailers, but at a price. Internet Retailer Magazine. Retrieved from https://www.internetretailer.com/2015/01/13/socialnetworks-deliver-retailers-price. Zollo, M., & Winter, S. G. (2002). Deliberate learning and the evolution of dyanmic capabilities. Organization Science, 13(3), 339-351.

224

Appendices

225

Appendix A1: The Learning Organization Survey (English version) 1. 2.

Please indicate the level of your position in the company Please indicate in which department you work

Member of Executive Committee Operations (IT, Development, Project Management

Manager

Sales/Marketplace

Employee

Human Resources

E-commerce

Finance

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

3. 4.

5.

6.

7.

Moderately Inaccurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

In this unit, it is easy to speak up about what is on your mind. If you make a mistake in this unit, it is often held against you. People in this unit are usually comfortable talking about problems and disagreements. People in this unit are eager to share information about what does and doesn't work. Keeping your cards close to your vest is the best way to get ahead in this unit.

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

Slightly Inaccurate

8.

Differences in opinion are welcome in this unit. 9. Unless an opinion is consistent with what most people in this unit believe, it won't be valued. 10. This unit tends to handle differences of opinion privately or off-line, rather than addressing them directly with the group. 11. In this unit, people are open to alternative ways of getting work done

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

Slightly Inaccurate

12. In this unit, people value new ideas 13. Unless an idea has been around for a long time, no one in this unit wants to hear it. 14. In this unit, people are interested in better ways of doing things. 15. In this unit, people often resist untried approaches.

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

16. People in this unit are overly stressed. 17. Despite the workload, people in this unit find time to review how the work is going. 18. In this unit, schedule pressure gets in the way of doing a good job. 19. In this unit, people are too busy to invest time in improvement. 20. There is simply no time for reflection in this unit.

226

Slightly Inaccurate

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Moderately Inaccurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

21. This unit experiments frequently with new ways of working. 22. This unit experiments frequently with new product or service offerings. 23. This unit has a formal process for conducting and evaluating experiments or new ideas. 24. This unit frequently employs prototypes or simulations when trying out new ideas.

This unit systematically collects information on: Highly Inaccurate

25. 26. 27. 28.

Competitors Customers Economic and Social Trends Technological Trends

This unit frequently compares its performance with that of: Highly Inaccurate

Moderately Inaccurate

29. Competitors 30. Best-in-class organizations

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

Slightly Inaccurate

31. This unit engages in productive conflict and debate during discussions. 32. This unit seeks out dissenting views during discussions. 33. This unit never revisits well established perspectives during discussions. 34. This unit frequently identifies and discusses underlying assumptions that might affect key decisions. 35. This unit never pays attention to different views during discussions.

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

Moderately Inaccurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Moderately Accurate

Highly Accurate

36. Newly hired employees in this unit receive adequate training. 37. Experienced employees in this unit receive periodic training and training updates. 38. Experienced employees in this unit receive training when switching to a new position. 39. Experienced employees in this unit receive training when new initiatives are launched. 40. In this unit, training is valued. 41. In this unit, time is made available for education and training activities.

This unit has forums for meeting with and learning from: Highly Inaccurate

42. Experts from other departments, teams, or divisions 43. Experts from outside the organization 44. Customers and clients 45. Suppliers

227

Please respond to each item in terms of how descriptive it is of your work unit. Highly Inaccurate

Moderately Inaccurate

Slightly Inaccurate

Neither Accurate Nor Inaccurate

Slightly Accurate

Often

Always

46. This unit regularly shares information with networks of experts within the organization. 47. This unit regularly shares information with networks of experts outside the organization. 48. This unit quickly and accurately communicates new knowledge to key decision makers. 49. This unit regularly conducts post-audits and after-action interviews

Please respond to each item in terms of how descriptive it is of your work unit. Never

Infrequently

Sometimes

50. My managers invite input from others in discussions. 51. My managers acknowledge their own limitations with respect to knowledge, information, or expertise. 52. My managers ask probing questions. 53. My managers listen attentively. 54. My managers encourage multiple points of view. 55. My managers provide time, resources, and venues for identifying problems and organizational challenges. 56. My managers provide time, resources, and venues for reflecting and improving on past performance. 57. My managers criticize views different from their own. Garvin et al., 2008. Original survey can be accessed at: https://hbs.qualtrics.com/jfe/form/SV_b7rYZGRxuMEyHRz First two questions have been added for this study.

228

Moderately Accurate

Highly Accurate

Appendix A2: The Learning Organization Survey (French version) Membre du comité de direction Manager Veuillez indiquer votre niveau de poste au sein de l'entreprise. Direction des opérations Commercial et Resources Humaines 2. Veuillez indiquer le département dans (IT, développement, et Marketplace lequel vous travaillez. projet) Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service

Collaborateur

1.

Pas juste du tout

Plutôt faux

Un peu faux

E-commerce

Finance

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

3.

Dans votre service, vous vous sentez assez à l'aise pour exprimer vos pensées. 4. Quand vous commettez une erreur dans votre service, il est probable qu'on vous la reprochera 5. Les personnes de votre service se sentent à l'aise pour aborder les problèmes et désaccords. 6. Dans votre service, les personnes sont désireuses de partager ce qu'elles savent sur ce qui fonctionne ou non. 7. Pour réussir dans votre service, il est préférable de ne pas exprimer ouvertement votre désaccord. Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service Pas juste du tout

Plutôt faux

Un peu faux

8.

Les divergences d'opinion sont les bienvenues dans votre service. 9. Dans votre service, une opinion n'est pas appréciée sauf si elle correspond à ce que les gens pensent déjà 10. Votre service à tendance à gérer les différences d'opinion de façon privée et cachée plutôt que les adresser ouvertement au groupe. 11. Dans votre service, les personnes sont ouvertes à des modes alternatifs de travail Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service Pas juste du tout

Plutôt faux

Un peu faux

12. Dans votre service, les nouvelles idées sont appréciées. 13. Personne n’est réceptif à une idée sauf si elle existe déjà. 14. Dans votre service, le personnel s'intéresse aux meilleures façons de travailler. 15. Dans votre service, les personnes préfèrent éviter des approches qui n'ont pas encore été testées. Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service Pas juste du tout

Plutôt faux

16. Dans votre service, les personnes sont trop stressées. 17. Malgré leur charge de travail les collaborateurs trouvent le temps de faire le point sur le fonctionnement du service. 18. Dans votre service, la pression liée au respect du planning entrave à la qualité du travail. 19. Dans votre service, les personnes sont trop occupées pour investir leur temps et leur énergie dans les améliorations à apporter.

229

Un peu faux

20. Il n'y a tout simplement pas assez de temps pour se consacrer à la réflexion dans votre service. Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service Pas juste du tout

Plutôt faux

Un peu faux

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Plutôt faux

Un peu faux

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Plutôt faux

Un peu faux

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

21. Votre service expérimente fréquemment de nouvelles façons de travailler. 22. Votre service expérimente fréquemment de nouveaux produits et/ou de services. 23. Votre service a un processus bien défini pour conduire des essais et pour évaluer des nouvelles idées. 24. Votre service utilise fréquemment des prototypes ou des simulations pour tester des nouvelles idées. Votre service collecte des informations de façon systématique sur : Pas juste du tout

25. Les concurrents 26. Les clients 27. Les tendances économiques et sociales. 28. les tendances technologiques. Votre service compare souvent sa performance avec celle des : Pas juste du tout

29. Concurrents 30. Organismes qui sont considérés comme étant les meilleurs dans leur catégorie. Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service Pas juste du tout

Plutôt faux

Un peu faux

31. Votre service pratique de vrais débats et des conflits productifs lors des discussions. 32. Votre service cherche à connaitre les points de vue divergents lors des discussions 33. Votre service n'accepte jamais de réexaminer les perspectives bien établies lors des discussions. 34. Votre service challenge des modes de fonctionnement fréquemment lors des discussions 35. Votre service ne tient pas compte des points de vue divergents lors des discussions. Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service Pas juste du tout

Plutôt faux

36. Les nouveaux employés reçoivent une formation adaptée dans votre service. 37. Les employés expérimentés reçoivent des formations périodiques et des mises à niveau dans votre service. 38. Les employés expérimentés sont formés lors de leur changement de poste dans votre service. 39. Les employés expérimentés sont formés quand de nouvelles initiatives sont lancées dans votre service. 40. Dans votre service, la formation est valorisée. 41. Dans votre service, du temps est consacré aux formations.

230

Un peu faux

Dans votre service, il existe un cadre pour apprendre et échanger avec : Pas juste du tout

Plutôt faux

Un peu faux

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Ni vrai ni faux

Un peu juste

Plutôt juste

Très juste

Souvent

Toujours

42. Des experts venant d'autres services, départements ou équipes 43. Des experts externes 44. Les clients 45. Les fournisseurs Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service. Pas juste du tout

Plutôt faux

Un peu faux

46. Votre service partage régulièrement des informations avec des réseaux d'experts au sein de l'organisme. 47. Votre service partage régulièrement des informations avec des réseaux d'experts externes. 48. Votre service communique avec rapidité et précision les nouveaux savoirs aux principaux décideurs. 49. Votre service procède régulièrement à des entretiens et des vérifications (post-audits) après la mise en place des actions. Merci de répondre à chaque question en choisissant la réponse qui caractérise le mieux votre service. Jamais

Pas souvent

Parfois

50. Mes managers sollicitent d'autres points de vue lors des discussions. 51. Mes managers tiennent compte de leurs propres limites concernant leurs connaissances, les informations qu'ils possèdent ou leur expertise. 52. Mes managers posent des questions pertinentes. 53. Mes managers sont à l'écoute. 54. Mes managers encouragent des multiples points de vue. 55. Mes managers accordent du temps, des ressources et des locaux pour identifier les problèmes ainsi que les défis organisationnels à résoudre. 56. Mes managers accordent du temps, des ressources et des locaux pour réfléchir aux performances antérieures et aux moyens de les améliorer. 57. Mes managers critiquent les points de vue différents des leurs. Translated from Garvin et al., 2008. Original survey can be accessed at: https://hbs.qualtrics.com/jfe/form/SV_b7rYZGRxuMEyHRz First two questions have been added for this study.

231

Appendix B: KIP, Performance, KM Strategies, and Types of Work KIP Type

What is High Performance

How is High Performance Achieved

What Needs to be Known

How Does Knowing Take Place

How is Knowing Applied

Examples of Types of Work

Introductory training and easy access to stored information through manuals, experts, and/or databases, couples with standardized procedures, scripts, and templates that document and demonstrate action sequences. Extensive study and training to master vast amounts of specialized expertise and to master numerous systematic and analytic search processes and problem-solving processes, accompanied by continuous learning, adaptation, and synthesis of new information

Work design based on replication and reuse of routines and work settings that facilitate storage and easy access to work practice information and continuous process- and outcomes-based performance feedback. Work design to use specialized expertise and work settings designed to facilitate systematic but flexible ways to transfer and share expertise; frequent education and reflection opportunities; and incident-triggered outcomeand process-based performance feedback Work design based on application of experienced based skills and work settings designed to offer related experiences, apprentice/mentor relationships, clear outcome expectations and feedback, and opportunities to hone and refine techniques Work design to use and leverage unique talents and abilities and initiate new challenges; work settings that facilitate communication, selfleadership, emergent processes, and social network development

Grocery cashier, bank teller, waiter, standardized manufacturing jobs, short-order cook in a fast food restaurant, mail carrier, medical records clerk.

Enacted information (low tacitness, high learnability)

High levels of efficiency, consistency, reliability, accuracy, low error rates, and predictability

Establishment and application of stable routines, application of best practices and template for action, extensive control systems, and clear knowledge of results embedded in work activities

Readily learnable information with clear cause-effect sequences that guide the design of routines; established procedures and action steps

Accumulated information (low tacitness, low learnability)

Products, services, and solutions that are customized, effectively meet situated client goals, accommodate unexpected parameters, and add value in unique and synergistic ways

Apprenticed know-how (high tacitness, high learnability)

Outcomes that meet specified quality metrics with a level of proficiency, craftsmanship

Application and adaptation of extensive and complex expertise through the use of logical and systematic analysis or the application of heuristic models across a wide array of factors and trade-offs to determine the most effective outcome according to externally established metrics Exploitation of tacit skills and collective capabilities developed over time; incremental process improvements based on personal experience, reflection, perception, and accumulated insight.

Extensive, complicated and detailed information that must be understood in relation to contingent factors and potential implications to provide a basis for situated analysis and task design, coupled with mechanisms for assimilating and integrating new information Learnable tacit skills, knowhow, and capabilities that are captured in muscle memory, experience-based interpretation and insight, and personalized routines

Talent and intuitive knowhow (high tacitness, low learnability)

Creation of ingenious, innovative, exceptional outcomes that define new sources of value creation or set world-class benchmarks or standards for products/services

Application and leveraging of exceptional human talent, creative destruction, invention, inspiration, and innovation

Unteachable, highly tacit, specialized, and high-value skills, know-how, and innate talents that are very complex, unexplained, ambiguous, and evolve with use

McIver, Lengnick-Hall, Lengnick-Hall, and Ramachandran, 2013

232

Extensive mentoring, learningwhile-doing, and experience in specialized but complex work settings using repeated experiential trials; extensive feedback to hone skills, enhance diagnostic capabilities, and develop a “feel” for the work Creation of opportunities and conditions that nurture innovative and creative minds, align diverse efforts, enable exploration, provide an array of resources and stimulation triggers, and encourage experimentation

Engineer, legal assistant, tax auditor, IT troubleshooter, financial analyst, building contractor, editor

Carpenter, electrician, plumber, mechanic, oil rig supervisor, radiology technician, pilot, social worker, military patrol leader

Artists, creative designers, elite athletes, creative writers, inventors

Appendix C1: Individual Interviews Real names have been changed to protect the confidentiality of participants



Individual interview 1: Yves, page 1 (December 2nd, 2016)



Individual interview 2: Simone, page 4 (December 8th, 2016)



Individual interview 3: Dominique, page 5 (December 9th, 2016)



Individual interview 4: Bernard, page 7 (December 14th, 2016)



Individual interview 5: Abel, page 8 (December 28th, 2016)



Individual interview 6: Karlotta, page 11 (January 13, 2017)



Individual interview 7: Richard, page 12 (January 13, 2017) o Total: seven: Five male, two female Reflective commentary in blue

Individual Interview 1, December 2, 2016 Yves – confounder, was in charge of sales and business development but is now one of two people in the communications/photography department. Q: Can you tell me about the history of company? It was founded by three different people with different sensibilities and experience—they were all working for a startup and decided to leave to set up their own startup. Eloise, whose interests, knowledge, and training were related to the Internet and the World Wide Web (studied Entrepreneurship); Yves was interested in photography and images (but who actually studied economics and business-related topics); and Mark who knew who how to raise money and set up a business (studied marketing and entrepreneurship). Q: Can you talk about the history of innovation in the company? According to Yves, the first innovation of the company was anticipating imminent changes in 2007 with the rise of Facebook, Airbnb, the iPhone, Uber, and user generated content, in general. They saw that the “two worlds” of e-commerce would change. They anticipated acceleration in the e-commerce sector. Brands tended to separate the stores from the web site. The founders of the Company understood that could join what Yves refers to as the “sales” site with the “image” site. The business model they invented was innovative because they offered a full-service ecommerce package—so clients would give them the entire responsibility for e-commerce, not just pieces of it. They charged a significant commission on sales (40%). According to Yves, this is similar to how franchisees function. So they were functioning almost like a franchise of the brand(s). Third, they wanted to accompany the brands operationally and strategically in their development of e-commerce.

233

Other innovations include a “responsive design” in the website conception so it will adapt to the size of the user’s screen; and offering the possibility for the consumer to find an item of clothing on line and then to reserve it in the store. Commentary: the barrier between e-store and the in-store boutique experience is getting less and less significant – the two are increasingly seen as part of an integrated experience. Q: Can you talk about the management philosophy and management innovations? Yves used the phrase “no sign of power” by which he meant visible signs of hierarchy. He stated this was beyond just having an open space. Managers do not have offices –in this way they “know what the other workers experience” physically in the workplace (for example, temperature). Management innovations that he wants to put in place in the next four months include having 80% of the workforce free to work wherever they wanted (i.e. no fixed work stations), and installing a café on the first floor for people to work in, and a library (I think he meant in terms of atmosphere) on the 4th floor, with the 2nd and 3rd floors for those who need offices (fixed work stations). Q: Provide another example of innovation. An e-commerce platform and system specifically for the fashion sector. They are working on developing this which is something they could license to brands. The objective would be to link a variety of functions together such as “new ecosystems”, “networks”, “multibrands”, “private sales”. It would link the client, the stock, and the promotional content. Apparently this would allow the client brand to print everything you see online (in the store). And it would somehow impact the pricing – prices would fluctuate according to various factors (location, season, demand?). Commentary: shift to product orientation Q: What is the knowledge required for these innovations? IT, knowledge of client needs. Mentioned smart identification by the site—to anticipate client preferences based on name, location (statistical analysis). Knowledge to accompany the client – strategic and operational. Knowledge of the brands. Knowledge of how the web functions – how to do marketing, the technology, the graphics. Commentary: some of this will be tacit in so far as employees are web users, hence second nature. Returned to idea of joining aspects of e-commerce that have previously been separate. On one side: image site (history and image of brand); on the other: sales. Yves wants to (and/or already has) put those together into one site with a different menu. Not quite clear to me what the improvement or innovation is to me. Example was something like this: Old Menu:

History of brand, creator

E-shop

New Menu:

Skirts, dresses, sweaters etc. History of the brand

Something about success – this is how it’s presented online: 234

Creator – style, inspiration

Product But the client is generally interested in function (like clothes for wedding, for work etc.) There are two types of sites now – the site described above and ones that are just functional. Yves wants to combine them. Q: Can you describe the company’s competitive advantage and threats? Ability to anticipate change. Example – smart phones are killing the coup de coeur (when you suddenly fall in love with a product in the store) – making shopping more rational. This and social networking is completely changing everything – vision and anticipation. Commentary – example of a potentially significant insight about the direction of e-commerce. Q: Threats? Shopify, a Canadian e-commerce company. Offers the possibility to make very cheap and effective e-learning platforms. There is no systematic analysis of threats at the company. Q: What are the company’s key competences you can’t imitate? Knowledge of brands, e-commerce expertise. With 30 brands, each week the company analyzes figures, looks at what works and what doesn’t, and makes continual improvements. With each site they try to do better than the previous one. Commentary: this is a key point relating to the firm-specific resources – accumulated knowledge through experience. Q: What makes for successful client relationship, how do you know when this occurs? Partnerships based on performance. Numbers. When the client speaks about you. Very good example: [Name of brand a} is growing. The e-store helped them improve their image and reputations. Bad example, [name of brand b]. Never agreed on objectives. Everything is complicated. Threats etc. Good example: [Name of brand c]. Been working together for 8 years (long contract). Bad example, [name of brand d] – they just didn’t understand how to sell. The Company could not explain it to the directors. But if they had sold well, it could have worked. Comment: it’s hard to understand more concretely why it didn’t work. Q: Can you describe the how decisions are made in the company? Decision making not clear – sometimes very centralized sometimes decentralized. This is disadvantage of the three founders. Yves describes his own decision making style as “yes but”; Eloise’s style is indecisive; and Mark’s style as very decisive and controlling. He compares Mark to Jeff Bezos. Brilliant ideas, very controlling. Also mentioned Elon Musk. Q: What constitutes high performance at the company? Best employees to their jobs but add something “extra”. Example of Simone brining experience from her previous position. Latifa helps with purchasing and put a box downstairs to collect donations. Someone who works in referencing said they should stop selling fur – might lose sales but would bring something social, meaning. Commentary: the notion of something “extra” beyond the job description comes up often. Employees are not expected to just fall in line and do what they’re told. 235

Individual Interview 2, December 8, 2016 Simone – HR Manager. Q: Can you talk about the company’s management philosophy? The mission is to faire grandir the team (personnel) and the client. This translates to something like ‘make grow’ or ‘facilitate growth’ in the moral/intellectual sense. For employees this means learning about themselves, taking risks and gaining skills. For the client (brand) this means accompanying them (the client) through the digital transformation, to think differently, reflect. Commentary: they have the attitude of a consulting firm. Q: What does the company do to implement this philosophy? They are investing 200K euro in training which will include emotional management, English language training, and time management. She described a training approach inspired by interval training in athletics (“a series of low- to high-intensity exercise workouts interspersed with rest or relief periods” – found on Wikipedia, reference to Vivian Hayward, 2006) – organized into 30 minute sessions. They are currently testing it. It’s internal in so far as one employee trains another so they learn from each other. Currently they are testing it on management, traffic, and time management. The idea is that the managers would become a like coaches. They would need to be trained in order to do this – so they become the experts. Their goal is for the (top) managers not to have any “production” role but only manage people. But then she seemed to revise this and said 30% operational, and 70% managing people. Commentary: this section of the interview is relevant to the Garvin survey – item about Education and Training. The emphasis on internal (employee-to-employee) training says a lot about the firm-specific nature of the activity. Are they being too internally focused? Q: What constitutes high performance? Entrepreneurial spirit, creativity, initiative, drive (what’s interesting is she used the English word for “drive”) The management philosophy since the very early stages of the company’s development has been to recruit young employees and to give them significant responsibility. A web developer is expected not just to fix the bugs in the system but to proactively identify them, seek them out and find solutions. Proactivity. Q: What is the “cause” of high performance? The attitude of the employee – sense of leadership, drive, proposes things, open to change. According to Simone, a manager creates opportunities for learning. Commentary: this puts the emphasis on the selection process in hiring rather than on training and development. Clearly when they are training their own managers, they need to train the managers how to create opportunities for employees to learn. Q: To what extent is decision making centralized at the company? According to Simone, much depends on what the employee(s) dares (oser) to take in terms of decisions. Everyone must take decisions. Example—they are digitalizing a lot of forms and 236

processes (or that’s their goal). Simone is delegating that responsibility to other employees— to listen to the sales presentations and participate in the choice of the service provider. Q: Can you give an example of a decision that you have taken – how did this transpire? Decision to give someone a raise. Simone asks her manager (Bernard, the financial director), who then asks Mark, the President, who responds that he trusts Simone’s judgment. Commentary: suggests that one person has considerable centralized power. Q: Can you give an example of a company-wide decision? Decision to move (appears to be imminent because necessary – goal is for April 2017) – question of where (in Paris? In Surburbs? Elsewhwere?). They are holding (or have held) six workshops to discuss which have (or do) include the employee representative (‘delegue du personnel’) – which is a formal role in French labor law. Management makes presentation apparently to the members of the workshop who give their input. Employees actively participate in the interior design of the open space. Simone also referred to what Yves spoke of – first floor café concept. Individual Interview 3, December 9, 2016 December 9, 2016 Dominique – head of commercial development and strategy He changed the economic model. Eight years ago, the company considered that clients didn’t know anything (about e-commerce) so the company proposed to do everything for them. The brand just supplied the merchandise and the company took a 40% commission (fixed rate). But now (for the past year), brands have their own digital services. 40% is too high. The offer is too “intrusive”. Plus there is a crisis in the retail sector. Dominique obtained the agreement from Mark, the President to change things. The rate of commission is now sliding from 11% to 23% according to volume (i.e. the higher the volume, the lower the rate). The added value of the company is its technical and marketing expertise. The additional services it offers are not its real added value: photography, logistics, transportation, after sales/customer service. There is a service fee rubric. These services are not their core business. He thinks the company is behind – not sufficiently innovative. But there is no real competitor in the sector – expertise in ready-wear retail e-commerce. They are agile in their various jobs (functions) and they accompany the client. They are slow on the “model” (not sure what he means here) and slow to test new things (technical things?). Agility is their main added value; they don’t have the best technology. Commentary: agility is both structural and part of the tacit skills of employees. Strengths They are the reference – reputation Their 36 clients Large size (relative for the sector) 115 employees No outsourcing Opportunities The sector lags behind in general when it comes to ecommerce Licensing the platform

Weaknesses High employee turnover Internal and external communication Not enough technology based sales (not entirely clear what is) Threats Consolidation in the sector Usure du métier (job fatigue, burnout) Time consuming nature of many tasks/jobs

237

Q: What kind of competitive intelligence is practiced in the company? There is no competitive intelligence (on a strategic level). They speak about it daily in his service but there is no formal process. Commentary: make link to Garvin survey. If there is a threat, it’s a threat for a “division” of the company but not for the whole company. Commentary: more of a product orientation means less of a services orientation (so less need for marketing, for example). Example of a threat: Far Fetch, which is a referencing system. Apparently it includes 1000 boutiques. A client visits the site, finds his/her product, and finds the closest store to buy it. Another example: API He does not agree (with Yves) that Shopify is a threat because it’s too “far away” – meaning they are huge – it’s a very simple system to use. Q: Can you provide examples of commercial success? Getting a new brand as a client. But it wasn’t according to the original offer (which was refused). They made a second proposal which was only based on traffic (e-marketing); which is to say ad words, etc. They were remunerated only based on increased ROI. This means a potential for an “up sale” (which apparently means selling other more expensive services later – I suppose he has the platform licensing in mind). This was their first sale according to the “new model”. Another example: A new client didn’t trust them. They thought the company’s offer was too expensive. Dominique apparently conducted an analysis of how much the client’s own ecommerce was costing them (which involved several partners) and convinced them that they could make a less expensive offer. They signed “with the idea” that the Company would increase sales from 1 million to 1.5 million. The applied the service rubric. This is an example of pedagogy and a financial (strategy?). Commentary: potential for inherent tension with services such as marketing or operations which have to keep the client happy after the deal is signed. Q: What causes success? Dominique developed a well-defined process which is the “client qualification”. It’s a rubric. This apparently is how the company is going to assess whether or not a client is a good fit. Commentary: example of adopting more explicit processes. They answered a need and listened to the client. They (or he) analyze(s) the client. He analyses growth, how many visitors on the site (not sure where he gets the info but apparently it’s publically available). In the initial visits to the potential client, they listen. They propose nothing. The process can be long. A contract for 4 years might take one year to sign. Often the decision maker (Client CEO) is in his/her 50’s or 60’s, an entrepreneur. They have to see things from the potential client’s perspective. It’s a question of method. Commentary: recurrent theme of seeing things from the client’s perspective. There are four sales people. If it’s worthwhile, (i.e. if there is a potential client which has an interesting problem to resolve), more than one (or all four) sales people will go to learn. 238

Commentary: example of a learning practice/culture – to what extent will this be captured in the survey? He mentions “clarifying the perimeter”. The “new model” avoids disappointment (i.e. expectations not met) and is supposed to transform cost centers into profit centers (apparently this refers to the service rubric). In the second example – creativity of the “deal”. That’s what works. Because “everyone knows the company is good, they seek to make the needs of the potential client meet the company’s organizational capacity. Agility. Commentary: creativity implies tacit skills. Individual Interview 4, December 14, 2016 Bernard – chief financial officer Q: What does faire grandir mean for the personnel? It means savoir être which is very linked to emotional intelligence. To be capable of managing and being managed. Referred to the “five pillars of emotional intelligence”. Training. Question of what is management: managing time, dialogue, attention to coworkers. Q: How do you know when it works? We are going to create a ‘barometer’ – a tool. A NPS (net promoter score) Q: Can you provide an example? Kristina in finance (anti-fraud) – creation of a trusting relationship. Doesn’t think you should shut people up in a technical position – they’ll never develop. Emphasizes adaptability. Don’t stagnate, change positions. Believes this can work on all levels of the organization. Q: What does faire grandir mean for the client? To be “immersed” in the client’s world, customer intimacy. You think with the client, listen to the client. Cultivate a relationship with the brands. There is a difference between savoir (know) and comprendre (understand). Today we are (too much) in the mode of executing services. Distinguishes between the product (platform) and the service (advice). Individual Interview 5, December 28, 2016 Abel - e-commerce manager Q: Can you clarification the relationship between marketplace, marketing, e-store and traffic? Marketplace is a startup activity, began as a (relatively) separate entity (department) one year ago. It includes 4-5 people. Madeleine (Group Interview 4) is in charge of marketing for marketplace. Marketplace was set up this way because it’s difficult to launch new projects within well established, well-oiled departments. New initiatives need space to breathe and to grow. Currently the company is considering merging the two activities (e-commerce and marketplace) because now marketplace is operational. Marketplace is one piece of the new business model. New business model means selling the new platform in 2017 – other services will become optional (i.e. not sold always as a comprehensive package). Abel’s team has 239

been the core of the offer up to now but in the future they (e-commerce) will be an optional part of the package. Marketplace means accompanying the brands like [name of brand c] who sell through their own stores, through department stores, and online. Marketplace is like a digital department store. The brand decides on the price, manages its stock, does its own photos etc. but benefits from the traffic on marketplace (such as Amazon’s marketplace). The Company is the intermediary between the brand and Amazon (for example) – they build the electronic connections. Q: How is commission dealt with? Commission is shared, recognition for good e-store management can be individual. But all employees get a % of turnover (sales). Q: What are the market segments represented by clients? In fashion there are about 7 levels (market segments) 1) 2) 3) 4) 5) 6) 7)

Top luxury Luxury Entry Luxury Premium Market Mid-Level Market Mass Market Discount The Company deals primarily with: Mid-level, Premium and Entry Luxury. Premium is a growing market, items cost from 200 to 400 euro. He describes this as “regular” fashion that looks like luxury.

Q: To what extent do project managers work with the sales team when a contract is being signed with a new client? In the best cases, there is good cooperation. In most cases, there is a separate discussion with the sales team, and then the deal is transferred to project without a well-defined process. Sometimes there is a delay between the signing and the transfer to project. For example with the brand, [name of brand e], 6 or 7 weeks went by before project started working on the deal. Why? Not entirely sure. Q: To what extent do brand managers and project managers work together? They should work together on a very regular basis. The project manager is there to help make it (the realization of the site) happen. Weakness – the role and the place of the project manager is not clearly defined. On the one hand, the brand manager has aspirations and ambitions (essentially the client’s ambitions); on the other hand, there are two or three project managers who are trying to “swallow” all the work. They are one or two weekly meetings (every Friday). There is a project committee consisting of the three brand managers (one additional person hired/position created which does not appear on the organizational chart) and the project managers and Abel. Commentary: could propose, like in the Edmondson study, to participate in these meetings.

240

Q: Can you talk more about Adrienne’s position – e-commerce position embedded in the client company (see Group Interview 4)? She is about to finish her mission at [name of brand c] – she was replacing the client’s own digital assistant who went on maternity leave so Adrienne was “externalized”. This was a test of a new way of working with the client and Abel considers this arrangement could also function the other way around – i.e. an employee of a client brand could come and work at the company. When asked what this revealed about the relationship with the client, he stated this was more a question of the individual attitude of the employee. It is more natural for younger employees to blur the lines defining who they work for, who pays them, where they work. Commentary: this is an interesting point because of the focus on the individual, not the client. Q: Can you talk about your vision of the company as a “hub”? According to Abel, the ambition of the company is to be a hub of knowledge and expertise. Marketplace is a reflection of this. They want to be a ‘go-between’ – where they create value is in the process. They do things faster than the brands would on their own. They provide and share expertise. Hubs = flow of people and “planes”. People from the brands will learn things here. They sell expertise in the field of e-commerce but Abel is not sure this is what the clients are actually buying. He thinks it’s more of a mindset, an attitude, the energy. A CEO (of a brand) will say that it’s really a pleasure to spend time in the company, that it’s a breath of fresh air. They feel (when they come to the company) that they’re in a bubble of innovation, optimism, and energy. They will continue working with the company for the energy and innovation. “We have this energy and enthusiasm but we (sometimes) waste it because of lack of processes”. Commentary: this might be the most significant metaphor. What are the tacit skills required to act like a hub? He puts the emphasis however on attitudes rather than skills – might be interesting to discuss the differences in Chapter 5. Q: What is high performance for a brand manager? There are three measures: turnover, quality of relationship with the client, and team growth (this is qualitative). Client feedback comes when Abel meets and/or speaks with the clients. They also have clients complete surveys twice a year about their overall satisfaction with the services. Team growth is a subjective judgment. Abel judges the energy of the brand manager, how he/she behaves, the fluidity of the work. Do they “move around”, do they “feed the teams” and are they “fed by the teams”. A bad example would be someone who sits at their computer all day with headphones on number crunching. Commentary: “feed the teams” metaphor. Why do you need to move around if you’re a brand manager? How does this related to tacit skills? Q: What do brand managers need to know to perform well? E-commerce expertise is the pre-requisite. They need an “entrepreneurial spirit”. Ability to work in an unstable (i.e. dynamic) high growth environment with loosely defined processes. They need a “commercial spirit”.

241

Q: How is high performance achieved? Informal coaching. Modeling (i.e. Abel is the example). Example – this morning he spent two hours with a potential e-store manager (new hire) to assess if he/she was a good match. They are building a training program. Q: How do you learn to be a good brand manager? Through practice and feedback from Abel and or Mark (the president/founder). Simone (HR) will help regarding the personality and motivation. Commentary: make link to McIver framework – practice and feedback. Q: Describe the tasks a brand manager has to accomplish. Mainly problem solving. E-commerce is complicated – lots of technical problems. Each Brand Manager deals with six or seven sites – the probability of a problem is high. The brand manager can add value by anticipating where the website will (should) go (develop). It’s too bad if 90% of their time is taken up by problems. They should be in discussion more with the client CEOs, should be building road maps for each brand which means where the brand should be in one year, three years, etc. And then breaking it down into specific actions for the next quarter. Seeking better execution etc.

Brand Manager

Pair of E-store managers

Two brands

Pair of E-store managers

Two brands

Pair of E-store managers

Two brands

Each Brand Manager manages about three or four binômes (pairs of e-store managers) who each handle two brands. How the e-store managers work together varies a lot from one pair to another . Q: What is high performance for the digital media manager (Group Interview 5, Amélie)? Build new traffic service programs using new media (social media) like Facebook, Snapchat, etc. Understand how this works. Build relationships with FB, but in place and measure initiatives. Abel will judge the performance, however, not just on whether she builds new programs. She should not be spending too much time in the building – needs to be spending time with the clients. In the future “we could measure turnover” generated by the position. It’s a business development position – sense of entrepreneurship = freedom and responsibility. This is a new position. Commentary: huge place here for tacit skills. Q: What needs to be known for this position? 242

Needs to understand not only e-commerce but also business (i.e. a P&L statement). Q: How do you learn how to do this job? Can be taught by example to a certain extent. But there is a lot of creativity involved – i.e. making choices, designing your job. Abel thinks a framework can be created to help this type of job develop – referred to the creation of marketplace. They have experience with entrepreneurial ventures. Commentary: on McIver’s framework, this is close to the artist or star athlete. Q: How does knowing take place? Creating opportunities, encouraging experimentation. Individual Interview 6, January 13, 2017 Karlotta - brand manager Q: What is high performance for an e-store manager? (E-store managers report to the brand managers) E-store earns high sales Q: How is this achieved? Competencies required are: 1. 2.

3.

Technical knowledge such as back office web. Processes are written. Karlotta doesn’t get involved in this. What she does is oral transmission Feel for/instinct for the business. (This is closer to the carpenter example from McIver). You have to have good reflexes which must be developed and honed from experience. You must have an ouverture au marché (open to learning from the market). This is what Karlotta really works on cultivating, transmitting to her e-store managers. Operational skills (the least technical) – she did not really elaborate on this.

The trend in her activity is towards more formalized processes, written processes. For example, the procedures for how to set up a site because when people know how to do this, and they leave, it impacts efficiency (knowledge leak). Karlotta must help the managers write these procedures. She mentioned that Stephen (Group Interview 4) was now working in a newly designed position which is transversal. This seems to be linked to the formalization procedures. E-store managers work in pairs – junior and senior, with the former dealing more with technical and back office issues and the latter dealing more with the business issues. For the senior e-store manager to be able to work well, he must have the support of the junior and vice-versa. Their roles must be very clear so they don’t get confused. Commentary: tacit skill here might be knowing what questions to ask.

Q: Tell me about what you have been doing recently. 243

She has spent time with each of the senior e-store managers under her responsibility to talk about which sites work well and why. If the brand makes bad decisions or manages their stock badly, the company can only help a little bit—she can only try to address what they (the company) can highlight/emphasize. She asks questions to her e-store managers such as: what are the best (selling) products? Are they well stocked and easily visible on the site? Did we invest in media well? E-store (in general) is taking more responsibility for this. She will have to help them. It requires some expertise from the traffic department but the e-store managers should be confident enough to challenge the conclusions of the traffic department. They need to understand the three aspects of the brand: product, traffic, brand. Individual Interview 7, January 13, 2017 Richard – after sales service manager Q: What is high performance in the after sales service department? Quantitative: speed, ability to take calls quickly, send e-mail messages. Qualitative: understand the processes of each brand, ability to listen, to resolve problems and conflicts Q: How to you achieve high performance? Training and accompaniment. Takes about one month of training and accompaniment to be operational. Richard will listen to calls, count the number of e-mails sent. With each new employee, he checks twice a day calls and e-mail messages sent. There is much written (manuals etc.) but he considers this insufficient. By talking to the employees he can transmit something. They can refer to each other – building links. He stresses it’s a hard job. Clients argue with you all day. If you build a good atmosphere, strong links, it’s your strength to face this difficulty. In terms of what’s written—all processes are written in diagrams, step by step—to outline the logistics, for example. But needs to be explained orally as well. Nothing should be learned only by writing because by orally transmitting, he can verify that it was understood. Commentary: example of how the tacit completes the explicit. You need both. Q: How do you learn to do this job? He described how he started on a 6-month internship contract. During this period, his boss left the company and Richard got promoted. Since he’s been in this position, the service has enjoyed the best results ever. Commentary: recurrent situation, people learn on the job, they are “thrown in” to it when a manager leaves. Q: Why are the results so good? Cohesion, team building. He cultivates this and invests in it. He implemented a more rational planning of lunch hours. His predecessor had experience in logistics and had some good ideas such as software to manage the calls. He spent a lot of time on this (perhaps why he left). Richard has built on the work of his predecessor but has focused mainly on mentoring. He shows the example. Each sales representative deals with four brands. Before he took over, it was seven brands. Now he starts off a new person with three brands. Afterwards he asks them which one(s) they prefer to work with. “Which brands make you dream?” Example, Hanna said [name of brand f]. But this is a complicated brand to deal with – has very high 244

volume. Sixty percent of the messages are in English. After two weeks she got the brand she wanted. Q: What does the company do with the info they glean from the final consumer? Currently, they do nothing. Richard has set up a new protocol of monthly reporting on the brand which includes quotes from the clients (final consumers) which he intends to send to the brands. Internally useful but potentially useful to B2B client as well. He hopes to achieve greater recognition of the after sales service department. His (secret) mission is to transform the service from something purely operational to something more “intellectual”. Q: What makes two brands different from the standpoint of after sales service? Volume. How demanding the customers are. Specific problems. Examples: [name of brand c] has very high volume. Many of the responses are canned: where is my reimbursement? etc. They have a special software program for this. Each brand has its set of canned messages. The brand [name of brand g] tends to get very specific messages about the materials which make up the clothing. The reps generally just search for this info online. [Name of brand h] has complicated logistics because of the way they stock their products (or don’t). Commentary: to what extent is the knowledge of the different brands and how they function tacit versus explicit?

245

Appendix C2: Group Interviews Real names have been changed to protect the confidentiality of the participants



Group Interview 1 : page 1 (seven people: 3 male , 4 female) – December 2, 2016



Group Interview 2 : page 5 (five people: 3 male, 2 female) – December 8, 2016



Group Interview 3 : page 9 (six people: 4 male, 2 female) – December 9, 2016



Group Interview 4 : page 11 (nine people: 3 males, 6 female) – December 14, 2016



Group Interview 5 : page 14 (seven people: 1 male, 6 female) – December 16, 2016



TOTAL: 34 participants: 14 male, 20 female Red text = reflective commentary Group Interview 1, December 2, 2016

1.

Rodrigo – web developer

2.

Latifa- human resources

3.

Eleonore – digital growth manager (a relatively senior sales position)

4.

Riku – business developer (more junior than Eleonore)

5.

Adrien – SEO

6.

Kristina – finance in the anti-fraud section

7.

Christian – developer

Q: Innovation – what does the company do that’s innovative, how is it innovative? Christian They are currently working on a new tool to help the client and the company manage the brand. They apparently have “carte blanche” on this specific project. There is a team of 8 or 9 people working on it with mainly technical skills including developers, graphic designers, integrators. This is related to the new business model, what AM refers to as the anticipated shift from services orientation to more product oriented. AM also mentioned the likely effect this would have on staff (i.e. needed expertise) – more technical. Adrien Said that there was a person in charge of innovation and “veille technologique” (competitive technological intelligence) who only stayed with the company for one year. According to Adrien, he left because his position did not produce any concrete financial result – no measurable ROI (implying the company’s management wasn’t ok with this lack of result). Also, apparently he proposed many projects that were never implemented. Adrien mentioned “sales tablettes” and something about transposing “the boutique” (not clear). This implies a limited vision of innovation and experimentation. Something is tried but if it doesn’t produce a tangible result, it’s discontinued. Related to survey items on 246

experimentation (21-24) and possibly openness towards new ideas as well– could help explain scores here (Traffic is under E-commerce department) Kristina In her job, a possible example of innovation (in the context of the company), is the internal newsletter to inform the other employees of the company about what fraud is and encouraging them to ask questions about fraud. Related to survey items on transferring and sharing information (44-49) Riku Proposing a different type of contract – a different business model for the client. (This is probably what Dominique, head of commercial development and strategy, was talking about in Individual Interview 3) Latifa Talked about the HR initiative “équation du bonheur” (happiness equation) focusing on three areas: time (for training, for example), attention (quality, human touch), and money (raises, Bonuses, etc.) Related to survey questions on training Q: What is the added value of the company (specific advantages)? Not sure who was talking here: Added value of the company is the expertise in the sector (ready-to-wear fashion) and reputation. All the competitors went out of business. Expertise in all the areas of e-commerce (sales, brand, technology, logistics, etc.) Q: Criteria for achieving a good relationship with the client: Not sure who was talking Open mindedness while listening to the interlocutor. Scope of action. Maturity of the brand (i.e. they work better with more mature brands), offering a calibrated (tailored) package. Ability to listen to the client is clearly an important (tacit) skill – something that can be learned and developed but tends to get overlooked in favor of technical skills. Christian and Eleonore They lost a brand a couple of years ago because: 

Bad relations with interlocutors



Site wasn’t good



There were a lot of interns working for the company at the time



Lack of communication

247

Communication was tense or “interrupted”. Clients don’t’ mind so much when you’re unable to make a deadline but they want to be told. There is sometimes a lack of transparency. The company is “opaque”. Example where information sharing is lacking both internal and external (link to survey questions) Christian Many people left the company during a period starting last April up until last September. Meanwhile, the sales team brought in new clients. Bad match. And with the project (referred to above), developers, integrators, and graphic artists were all very busy. It was a management crisis and it was “structural”. The problem of mismatch between new clients being brought in and the ability to meet their needs expectations is a recurrent one. Q: What are the competencies required, particularly for team work? Communication, listening skills Adrien There was someone hired recently who was supposed to render the work lighter for the director of operations. Many people quit. Thanks to him (or to this situation), “we understood there was a problem”. This person left after a month. Then two other people left. Rodrigo Forming teams is difficult (i.e. this is in progress but the high turnover is a challenge, plus he is new). Adrien People who were recently hired do not get sufficient accompaniment. The company is so much in the operations that they can’t think. Not all teams function well. Lack of action to follow up words Lack of training and time for reflection (link to survey scores). Riku: Rigor Christian Graphic designers integrators and developers work together. Need good communication, “lacher-prise” (ability to let go if something isn’t working), patience. Interesting tacit skill which, according to him, is necessary for team work. Katrina The fraud department is small. They listen to each other. Even to the intern. Example of dealing with a brand where they notice a lot of fraud in consumer purchases. They put in place a security code protocol in the transaction (consumer’s bank sends a code by text 248

message which consumer must use to validate transaction). This security protocol is validated by Bertrand, the financial director. Then it’s up to the Brand Manager to deal with the company (implies they would ask the brand/client or just impose it – not sure). Example of how information is gathered and moves up. Q: How are decisions made? How do employees learning from one another? Eleonore The commercial strategy is very centralized. Decision making takes time, and is rarely well communicated. For the past year, this has been very blurry. Lack of clear decision making processes is a recurrent theme. Adrien Speaking about the “Projet Charlotte” – collaboration of working groups of 5 or 6 people during the summer of 2015. “Why so much turnover? What can be improved? We don’t learn from our mistakes.” From one site to another. But what happened? 1) new contract; 2) new project It appears that one project fell apart and/or was replaced by another project. Not learning from mistakes – knowledge is not shared, not explicit, “leaks away” with high employee turnover – check article which deals with this exact topic. Christian No manual. Nothing written about how to make a good site. No documentation. Example of someone who made a manual but only he understood it. They reworked it. It explains how many steps for the brand to function in how much time—ended up being useful. Learning here becomes explicit, stored and retrievable. Adrien This is necessary for payments system. Because they have made the same mistake in setting up a payment system from one site to another. Because the person who knew how to do this left. Rodrigo Rodrigo must work on a lot of stuff that isn’t really his responsibility – i.e. when employees don’t know how to do things because they haven’t been trained, he has to help. Some people continue to make the same mistakes. Lack of explicit, retrievable knowledge. Riku There is a friendly (chummy) atmosphere – part of the open space. If someone is right next to you, it’s easier to ask your neighbor. People are always supposed to be available. This is a challenge.

249

Example of what Wilson et al. (2007) refer to as “paths” (p. 1053). “…the preference not only affects the group’s current learning but also affects group norms about which storage bins it consults in the future.” The “path” to Rodrigo is stronger than the “path” to a more formal set of instructions. Is this learning good? See “dysfunctional routines” – second to last para. Page 1054 Q: What is the line of authority for decision making? What is the management structure of the company? 1 brand manager for 6 or 7 e-stores. 20 to 30 people work for e-store.

E-commerce Manager Brand Managers E-Store Managers E-Store Assistants

Group Interview 2, December 8, 2016 1.

Margaux – supply chain logistics manager

2.

Nathalie – project manager in the finance department (anti-fraud)

3.

Jean Baptiste – web developer (back office)

4.

Geoffroy – lead IT architect

5.

Raymond – head of traffic and CRM

Q: Describe your jobs and areas of responsibility. Raymond Creating commercial offers—designing a new framework for this. It has to be customized designed for each client. Reporting, KPIs both internally and for clients. Managing a budget. Q: What do you need to know for this? Raymond You need a vision of the market, market and technical expertise. Proposition de valeur (value proposition) means “accélérateur du business” (business accelerator). “Vison of market” – complicated to develop and to transmit. Interesting topic to explore further.

250

Geoffroy He changed jobs a year ago. He was a developer for 4 years. That was just execution – was boring. New project – review all the servers. The company hosts 20 sites. This is a big challenge. His job is less routine now. Management is new to him. Q: What do you need to know for your job? Geoffroy Key issues with respect to the brands, key marketing dates (when sales happen, private sales, events, etc.), when sales tend to peak, understanding of certain technical constraints. Challenge –since the sites run 24/7, when you make infrastructure changes, you have to get the client to approve the dates of when the sites might be temporarily down. Project management challenges—you depend on 10 other people. Have to manage the operational plus the management. Difficult to be an expert and a manager. He’s in the middle – between higher up managers and operational people Reminds me of one of Nonaka’s theories about the importance of middle managers because they connect two worlds. Margaux You need accompaniment, presence (not sure of whom) Jean Baptiste Working on the Batlik study/project. Two weeks of study of the needs and the conception. Tickets have been written. The realization is solitary work. You just develop. Knowledge needed – web development, technical reflection, logic, link between technical and user. Nathalie Manages fraud issues for all the sites. Her routine: manage all the orders which are suspect. She has to individually examine them to analyze – it’s essentially a behavioral analysis. Knowledge needed is primarily knowledge of brands and clients. Ability to identify (or have experience with) certain patterns of behavior. She has to improve the platform and payment systems. She has to render the department and fraud issues visible to the rest of the company – working on a newsletter. Patterns of behavior have to be observed. Example: with one brand, virtually all the fraud occurs in Marseille (geographically). Identifying characteristics – small purchases. Whereas with another brand, the typical fraud case might involve large purchases. There are certain delivery modes which are suspect (with certain brands) as well as certain e-mail addresses. Knowledge is gained through experience. How to make something explicit, shareable, retrievable if it’s constantly evolving?

251

Margaux Works closely with another person. She described routine tasks which involve tools and transmission of information. She has to transfer many messages (which is dull) because their service (logistics) is the entry point for many different issues and interlocutors. She described more complex tasks as business analysis, diagnostic and recommendation. She described the need to break out of the daily routine, to use her brain differently, to change perspective—“we get numbed by our daily tasks, lose priority, it’s the malaise of our society.” She wonders how to put in place things which can help maintain malleability (in thinking and acting). Different kind of knowledge: savoir être. Interaction with the client and colleagues. “We don’t’ listen enough or communicate enough.” She wonders how improve this area of competence (which she puts under the heading savoir être), related disciplines. Identified need to develop these competencies – ability to interact with client. Q: What is high performance? Margaux A clear, precise, and feasible objective is necessary for high performance in her job (logistics) Raymond Turnover. It’s quantitative (CRM and traffic analysis) – numbers of visits on the site, for example. Must explain to the client and internally. Problem identification, and global understanding of problems are necessary for success. It’s like understanding how a car motor works. Tacit skills necessary to develop a “global understanding”. Metaphor of car motor. Margaux It’s a frame of mind. I take initiatives. I’m not afraid to fail. Everyone agrees here accept Nathalie. Link to survey questions on experimentation and psychological safety. Nathalie Many things are stuck. She is confined to her tasks. She does not have the means at her disposal to take the initiative. See above – finance may be more structured, risk averse. See Edmondson’s work on variegated learning – finance (at least on Nathalie’s level) may not allow for double-loop (radical) learning.

252

Q: What are some examples of success/high performance? Margaux Putting in place a process, a tool. Knowledge needed: savoir être being able to see things from the other’s perspective. In the beginning you drown in the details. Key word: simplicity. This comes from exchanging with others. Evidence of tacit knowledge – shared by sharing experience. Raymond Create fertile ground which favors chance of success. Elements include good communication and being “open to the market”. Success is not individual (except maybe for sales people). You have to open the hood of the car. You need empathy, comprehension, sharing. Returning to the metaphor of the car motor. Emphasis on soft skills—communication. Q: What is competitive intelligence? How is it practiced? Geoffroy They have a tool which all employees can use: elCurator, which is an aggregator. Employees suggest things (to the company) but nothing is applied Margaux we do it (intelligence) naturally Geoffroy Should be done by each person. He insisted this should be more or less automatic (or “natural”) for people working in e-commerce. “It should even be an interview question before you get hired. Candidates should be asked how they do it. But we can’t do anything by ourselves unless we just implement things without telling anyone.” Example: new method to deploy source code on the server. He spent the weekend studying this. Then he has to “sell it” to his manager to “sell it” to the marketing team and then to the executive committee. Usually it’s the client who wants it or the executive committee which has to claim it as their own (or feel like it was their idea). This example related to survey results on Psychological Safety, analysis, and other items. Suggests that bottom up doesn’t always work well. As per Edmondson’s work, egos get in the way. Discussion about redoing the checkout system three times with the same result. This was done just because a manager said it should be done. The management needs to prove they’re right. It’s because of the egos of the people on the executive committee. Q: Describe the relationships you have with clients. Geoffroy There are two types. One is young and they listen to you. They don’t’ tell you what they want (presumably because they don’t know). The other is more mature. They think they’re experts and tell you what they want.

253

Raymond There is a type of client which is in between the two types. These clients are starting to know what they want. Margaux Need to put yourself in the client’s shoes and explain things in a simply way. Listening skills. Group Interview 3, December 9, 2016 1.

Mariane – assistant e-store manager

2.

Matsu – web designer

3.

Benoit – junior e-store manager

4.

Jeremy – junior e-store manager

5.

Joseph – Web developer (back office)

6.

Mathieu – Web developer (front office)

Q: Define success. Mariane It’s when you promote the right product at the right time. This involves analyzing traffic, sending newsletters, “retargeting”, SEO, verifying the stock, good “commercial operations” (meaning a specific event like a private sale). For this to happen, the brand has to give you the stock on time, the web designers select the right look, they send out the newsletter, the web integrators set up specific pages correctly. Recurrent theme for organizational effectiveness – how the different parts work together. Requires a very high level of cooperation and implies need for much info sharing. Michael Reaching an objective, well-defined objectives, resolving bugs when the site comes out. Benoit When the brand makes money. There is effective communication from creation to integration and development. Mariane If it doesn’t work, its’ because of lack of coordination (internal and external). Example of when there were external service providers involved who were not ready on time. Benoit Used the term “free style” (in English) apparently referring to when they have to work under tight deadlines and improvise. A kind of metaphor for tacit skills. Here we can ask though if this type of learning is good. Is the company developing bad habits?

254

Joseph Success isn’t “I have finished my task”, it’s “the client is happy”. When private sales break records (exploser le record). Work together Mariane We (the company’s sales team) sell a date (by which the site is supposed to be ready). It’s not always feasible. Delays accumulate. Bugs (in the site) have to be resolved later. Joseph and Benoit Spoke about the esprit d’anticipation, working under pressure – reaction rather than anticipation. The company does not anticipate enough. Mariane Her most interesting work is with clients on the content of the site rather than just on commercial operations (meaning short term events like sales). She mentioned working with clients who are confident in their brand and don’t want to just discount the prices. This suggests there might be some relevance in applying or developing a client life cycle model. Related to accompanying the client. Could this help in terms of a shared approach? Benoit We work on the quantity, and aren’t happy with the quality. Like a machine. The only sense of well-being we have is when we have managed a problem. We only have team meetings when there are problems. Matsu States that in her department (web design), they have regular meetings. Edmondson – work groups function very differently (and this impacts learning). Raises an interesting question: how can work groups/departments learn from each other (i.e. not individual-to-individual learning but group-to-group learning). If someone from web design sits in on a marketing meeting (Jeremy and Benoit) to compare meeting management. Q: How do you say ‘no’ to a client demand? Benoit You have to demonstrate that it won’t work. It takes charisma and diplomacy and good arguments. You have to do analyses. Q: What is innovation (in the company)? This discussion did not really address innovation as hoped; it was really about perceived lack of innovation Benoit Wants to have training sessions led be experts. Or by the brands, or with the clients.

255

Matsu Spoke about the vivre ma vie (live my life) HR initiative which allows you to pair up with someone temporarily to see how others work in other departments. Joseph If we want to change our technology, there are 24 sites which have to be changed. Also when the company went from 20 to 24 brands, there was more work, so they need more people. In terms of technological innovation, he says there isn’t any. Benoit Since the sites run constantly 7 days a week, it’s difficult to do the work necessary to improve them. Group Interview 4, December 14, 2016 1.

Laurene – web designer

2.

Leonard – financial controller

3.

Stephen – brand manager

4.

Javier – SAAS product manager

5.

Maude – project manager

6.

Adrienne – e-commerce position embedded in a client company

7.

Madeleine – account manager for marketplace (Marketing), formerly assistant E-store manager

8.

Celeste –operations manager for marketplace

9.

Solange – HR (recruitment)

Q: How did you learn to do your current job? Laurene Most of what she does (web design) she had learned before her current job (i.e. how to use web design programs). She said she learned about 20% of what she does on the job which she defined as the “form rather than the content” – the specific way of working at the company. Apparently she was paired with someone when she first started (a sort of mentor) to learn this. “The way of working” = tacit learning. Transmission: mentor. Leonard Got his foundation at school (finance). On the job, what he learned which is specific to the company is the business model, and what is expected in terms of summarizing and presenting the information. His manager told him what to emphasize.

256

Stephen You don’t learn the strategic parts of the job at school. He uses the Magento e-commerce platform (a customized version). He mainly learned by himself but his manager taught him a little. There are “so many possibilities” with this platform that you can’t teach it all. Reflection on KM in an environment where IT is key. Javier It depends on what job. For him it’s a lot of trial and error. He is self-taught for SaaS. To learn the business side, on the other hand, this comes from exchanging with others on a continual basis. For the management part of his job he had a prior training course (not related to the company) on management based on theater improvisation and situational role play. Maude Project management—she already had significant experience when she came to the job. When she was hired, her manager was on paternity leave so she had to be quite autonomous. but stressed she learned a lot from exchanging with other members of her team. She leaves a written trace of what she does to make sure certain things (processes, how to do things) are recorded. Recurrent theme: someone ends up in a job when someone else leaves, was never replaced, or never really there at all. She is moving certain types of knowledge into a more explicit format. Adrienne She had to learn a lot on the job although she did receive some company training. She said this training wasn’t operational. Like for Magento, she had to learn by doing. She was actually placed in a job at a client company (i.e. embedded)—apparently she covered for one of the client’s employees who was on leave. She had to be really proactive. Exchange with teams. She stressed that an interlocutor was an opportunity to learn. Madeleine She started as an intern. Her manager left. She did not get explicitly trained. She had no experience with digital technology. Apparently she moved from e-marketing to marketplace – where there are many external partners. Had to learn an integrator system (Lengow) – learned with manager. She learns constantly and states that it is difficult to learn e-store brands by yourself because they are complex and information rich. Once again – e-commerce is highly social at this company, how people learn – from others, not from manuals. Dynamic, evolving environment. Ceclie She is the operations manager of marketplace. Her job was created around her, to a certain extent, because it didn’t exist before. She had to learn Lengow which centralized everything—one interface to send information to different sites internationally. She said that vocabulary is different from one company to another. She went from project management to operational management, and says she her job is 60% operational and 40% project currently 257

(while stating it should be the other way around). She learns thanks to exchanging with teams. Gathers info. What is lacking is process but processes are emerging. They create processes. Implies that in her department, they communicate more than elsewhere in the company. She mentions lack of interdepartmental communication. Reminds me of Christensen’s statement about companies moving from just people knowing things to more formal processes. Processes are merging. But lack of communication between departments again—related to survey questions about knowledge sharing. Solange Her job is totally different than in her previous company where she was more autonomous. There is more teamwork at this company. There are weekly team meetings. She had to learn how her actions impacted others. She stressed that the current HR department was only 10 months old. Q: What does faire grandir mean when it comes to the client? Javier It means how to aid the company (client) boost its business more autonomously, in more intelligent ways. Example: if the client had realized themselves that there was a rupture in stock and had anticipated better, instead of realizing it at the last second. To use the bicycle metaphor – the idea is to teach the client to ride the bike himself while the company offers advice. Bicycle metaphor. Madeleine Madeleine argued this works for new brands/clients, whereas some of the clients that have been with them for a while expect them to do anything and everything for them Q: How do you deal with that? Adrienne We need to share with the client what’s difficult. Internal communication is complicated because the project doesn’t “come out”. Not sure what she means by the second phrase. Related to information sharing and transfer. Maude Mentioned the ability to realize a project (implying it wasn’t taken into account). Javier Lack of communication – if marketplace wants to bring in a brand (i.e. a new client) – sales people make promises that are impossible to keep.

258

Laurene We always make compromises. We bargain. We have someone whose job it is to do that. Take care of emergencies. Example – change how a website functions, change its image. Retour en arriere (regression). Flexible. Interesting example – they have explicitly recognized that problem solving is such an issue that it has become someone’s job to deal with such issues. Is this good learning? Javier There is tension between the “industrial/factory” attitude and attention to quality (which he associates with marketing) in each service. Sales has to sell. Difficult to find common ground. Central tension at the company Q: Do you learn from your mistakes? Madeleine Makeplace needs e-marketing for information. Needs information for a sale (i.e. to organize a sale). Half the brands are in marketplace. Lack of information, lack of processes. Mentioned the need to centralize Excel spreadsheets, for example. Concrete proposal for making info explicit, retrievable across departments. Celeste We must anticipate a lot in marketplace. The next day—update. Example. Prices for sales (they lacked this info). They ask a month in advance. Implies that marketing is more last minute. Stated there are nine brands in marketplace. New business model is driving the need for different organizational skills: anticipation. Stephen Game theory. Coordination en amont (upstream). Javier The project manager has a transversal vision of things. She makes the file (specifications) – will find the necessary info. She knows what’s realistic because she is close to the practice. The “internal ambitions” of the company make this difficult (meaning someone in the company, possibly the president, thinks bigger than present ability to meet client needs). Key question: to what extent does the project manager usually work with the sales team when the original agreement is being established?

259

Group Interview 5, December 16, 2016 1. Jocelyne – e-store manager 2. Liliane – logistics and transport 3. Lucie – e-store manager (senior) 4. Khalid – brand manager, head of development for marketplace 5. Amelie – digital media manager 6. Andrea– traffic manager 7. Maxime – web designer Q: What does your typical day consist of? These questions followed consultation with the dissertation supervisor because of the inherent difficulties involved in revealing tacit knowledge. He suggested modifying the questions – trying to get people to talk more about what they do. Jocelyne She launches sites, works with two brands, works in binome (with a peer/partner), she reports on figures, commercial (sales) activities, follows up with tasks, tries to resolve bugs, uses a software program, talks to the brands on the phone, scheduling, planning the next steps for the season. Every day—site maintenance, import photos, manages stock etc. Logically then she is a junior e-store manager because she is clearly dealing with operations. Liliane Deals a lot with ADS (logistics service). At each step she verifies if all is well. Is the stock received and “integrated”? Measures the “service rate”, packaging. She negotiates with the transporter. Billing. Projects linked to the transport, physical flow of goods and digital “flow”. Makes sure the stock “goes up” to the site. Manages the images. Deals with discrepancies resulting from errors. Deals with 18 brands. Procurement is her boss’ responsibility. Lucie Deals with two brands. Usually there is a junior e-store manager who deals with operational stuff and the senior (she is senior) who deals with analyzing and following the market. They don’t divide according to brands but according to function. Khalid Has two areas of responsibility. He tests markets for brands, and he maintains the “ecosystem, which means negotiating the brand’s market position—in the large sense. He works on its visibility. Commercial development. He does quantitative studies—before and after assessments. Amelie Works on the brand reputation across digital channels. It’s a new position. She works on social networks like FB and managing partners. She conducts analyses to define a strategy 260

for each brand. E-store might conclude that a brand isn’t working and needs to improve. She will seek new ways of developing the brand, new projects, negotiate new partnerships. The user creates content (for example), and a company might collect the right to use these things—photos posted by users. She will try to negotiate with this company and with 30 brands, she has much leverage. She follows traffic and expenses. To what extent is this example relevant to tacit knowledge? Q: Clarification on the number of brands. The company manages e-stores for 20 brands and earns commission on sales. The rest are marketplace – (i.e. the other service model?) She listed marketplace, media and e-store as separate services in the company. Andrea Looks for new users, advertising and publicity, ways of boosting performance, ROI, increase traffic. She works on digital referencing and display. There are three people with the same job who share 11 brands. She has two “internal portfolios” and works on specific strategies (external), invests a percentage in customer acquisition. Maxime Sets up sites, meets clients, creates e-shops. She has constraints related to how she must cooperate with e-store marketing. Q: Tell me about a recent problem and how you resolved it. Lucie Spoke of a communication problem. There was a technical migration which apparently had bad effects on the client and were not anticipated. The client wasn’t expecting it (“a subi la situation”). She spoke of a “second step very oriented towards the client”. “Identification” (not clear if she was talking about the solution here). Q: Who do you speak with when you have a problem? Jocelyne She will speak to her partner. IT migration example – she will talk to the web integrator (Geoffroy). She was in contact with the brand. The brand couldn’t do certain things because their service provider apparently had constraints. And then she spoke with Geoffroy. This is very specific (i.e. not a process). If there is a bug in the program, she makes a ticket, and it is assigned. Khalid The brand doesn’t calibrate well (align cost with revenue?). Example of a brand in bankruptcy which was acquired by a Turkish company. The interlocutors were not great. Khalid lacked info about the products so couldn’t sell. So they made some basic assumptions—estimate roughly what is in the warehouse. Sales weren’t covering the costs. What does he do? He speaks to the brand. If that doesn’t work, there are punitive measures—i.e. send a bill to cover losses (for example). This is probably a somewhat typical case for the sector but it’s new for marketplace. Usually they can “mutualize” (i.e. pool/share) stock for marketplace but 261

here they can’t because it is too channel specific. There is no process to deal with this. He speaks with his manager. They have weekly meetings to deal with stuff in general. Khalid has to write a report. Interesting example – they have to be very proactive and flexible. Q: If you want to explore new ideas, who do you talk to? Maxime How to improve a site—she discusses with integration and development. When it comes to a new brand, her work impacts others. Is there a formal process for this? Q: How do you constitute a group? Bensimon (brand), Maxime (graphic design) and other team members. It’s the project management department who will do this. A Project Manager is assigned to the team. Difficult deadlines. Adrien (SEO) will get involved (but not technically part of the team because they don’t need him continuously). The project manager is central – relationship with the brand. Q: Describe a problem and how you resolved it (or one in progress). Liliane The system which managers the orders—many files are not integrated. Clients are receiving orders for free (among other problems). Client services noticed many orders on standby (that’s how she became aware). She will speak with the development manager (or was this her own Manager, which is to say Margaux?) He/she told her to conduct an analysis, a cost estimate, and to send an e-mail to all concerned. Her manager helped her to take concrete actions because she didn’t know how to react. To what extent is this tacit knowledge? Knowing how to respond to a crisis. Amelie Has a problem with a major brand. Every year they agree on the turnover target. One of the top managers (her boss) had an Idea: if they do advertising on TV, they will boost sales. They didn’t really study this—i.e. the potential ROI. They simply said ROI would be “crazy” (i.e. great). But it wasn’t so easy, not a simple model to work. Shared risk. Media: advertising on TV is supposed to generate traffic. The client communicated its amazing forecasted figures to its holding company. They got the media business from another company (not quite clear how this worked). Amelie analyzed how much to spend: 100K. Her boss spoke to the client (these appear to be informal meetings), then the brand manager, then finance. “We sold them a dream”. She addressed this with Raymond in an informal manner. Example points to lack of formal processes when it comes to estimating costs and work involved and the potential consequences.

262

Appendix D: Complete Scoring Data Sheet for Learning Organization Survey Learning Environment composite Liberté d'expression et droit à l'erreur Question Très juste Plutôt Un peu Ni vrai ni Un peu Plutôt Pas juste juste juste faux faux faux du tout 3 20 17.16 14.3 11.44 8.58 5.72 2.86 4 2.86 5.72 8.58 11.44 14.3 17.16 20 5 20 17.16 14.3 11.44 8.58 5.72 2.86 6 20 17.16 14.3 11.44 8.58 5.72 2.86 7 2.86 5.72 8.58 11.44 14.3 17.16 20 100 Total possible Appréciation des différences 8 25 21.42 17.85 14.28 10.71 7.14 3.57 9 3.57 7.14 10.71 14.28 17.85 21.42 25 10 3.57 7.14 10.71 14.28 17.85 21.42 25 11 25 21.42 17.85 14.28 10.71 7.14 3.57 100 Total possible Ouverture à l'égard des idées nouvelles 12 25 21.42 17.85 14.28 10.71 7.14 3.57 13 3.57 7.14 10.71 14.28 17.85 21.42 25 14 25 21.42 17.85 14.28 10.71 7.14 3.57 15 3.57 7.14 10.71 14.28 17.85 21.42 25 100 Total possible Temps pour la réflexion 16 2.86 5.72 8.58 11.44 14.3 17.16 20 17 20 17.16 14.3 11.44 8.58 5.72 2.86 18 2.86 5.72 8.58 11.44 14.3 17.16 20 19 2.86 5.72 8.58 11.44 14.3 17.16 20 20 2.86 5.72 8.58 11.44 14.3 17.16 20 100 Total possible Learning Processes Composite Expérimentation 21 25 21.42 17.85 14.28 10.71 7.14 3.57 22 25 21.42 17.85 14.28 10.71 7.14 3.57 23 25 21.42 17.85 14.28 10.71 7.14 3.57 24 25 21.42 17.85 14.28 10.71 7.14 3.57 Information collection 25 16.66 14.28 11.9 9.52 7.14 4.76 2.38 26 16.66 14.28 11.9 9.52 7.14 4.76 2.38 27 16.66 14.28 11.9 9.52 7.14 4.76 2.38 28 16.66 14.28 11.9 9.52 7.14 4.76 2.38 29 16.66 14.28 11.9 9.52 7.14 4.76 2.38 30 16.66 14.28 11.9 9.52 7.14 4.76 2.38 Analyses 31 20 17.16 14.3 11.44 8.58 5.72 2.86 32 20 17.16 14.3 11.44 8.58 5.72 2.86

263

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

2.86 5.72 8.58 20 17.16 14.3 2.86 5.72 8.58 La formation 16.66 14.28 11.9 16.66 14.28 11.9 16.66 14.28 11.9 16.66 14.28 11.9 16.66 14.28 11.9 16.66 14.28 11.9 Transfert et partage des informations 12.5 10.74 8.95 12.5 10.74 8.95 12.5 10.74 8.95 12.5 10.74 8.95 12.5 10.74 8.95 12.5 10.74 8.95 12.5 10.74 8.95 12.5 10.74 8.95

Leadership qui renforce Toujours Souvent 50 51 52 53 54 55 56 57

12.5 12.5 12.5 12.5 12.5 12.5 12.5 2.5

10 10 10 10 10 10 10 5

Parfois

11.44 11.44 11.44

14.3 8.58 14.3

17.16 5.72 17.16

20 2.86 20

9.52 9.52 9.52 9.52 9.52 9.52

7.14 7.14 7.14 7.14 7.14 7.14

4.76 4.76 4.76 4.76 4.76 4.76

2.38 2.38 2.38 2.38 2.38 2.38

7.16 7.16 7.16 7.16 7.16 7.16 7.16 7.16

5.37 5.37 5.37 5.37 5.37 5.37 5.37 5.37

3.58 3.58 3.58 3.58 3.58 3.58 3.58 3.58

1.79 1.79 1.79 1.79 1.79 1.79 1.79 1.79

Pas Souvent 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5

5 5 5 5 5 5 5 10

Jamais 2.5 2.5 2.5 2.5 2.5 2.5 2.5 12.5

Adapted from Garvin et al., 2008.

264

Appendix E: Complete Breakdown of Benchmark Scores Compiled by the Survey Authors

Garvin et al., 2008.

265

Appendix F: Table of Combined Survey Results Form Number Survey Question/Section

AVERAGES MIN 3 4 5 6 7

Liberté d'expression et droit à l'erreur 8 9 10 11 Appréciation des différences 12 13 14 15 Ouverture à l'égard des idées nouvelles 16 17 18 19 20 Temps pour la réflexion Learning Environment composite 21 22 23 24 Expérimentation 25 26 27 28 29 30 Information collection 31 32 33 34 35

18.31 11.03 16.00 17.49 14.16 76.99 19.47 18.71 15.36 19.55 73.08 23.30 19.81 20.48 16.58 80.16 10.01 14.02 9.17 9.26 9.80 52.27 70.62 17.21 17.68 13.26 14.45 62.61 9.12 12.81 10.77 12.24 10.83 11.05 66.81 14.03 14.98 14.64 14.84 15.79

Q1

Median

Q3

MAX

51.48

68.62

78.66

87.89

97.16

39.27

60.69

74.98

85.70

96.42

46.41

71.41

80.14

92.84

100.00

14.30 43.94

34.32 62.75

51.48 72.08

65.78 78.89

97.16 91.98

28.56

53.55

64.26

74.97

89.28

30.94

60.10

71.40

76.24

90.44

266

Analyses 36 37 38 39 40 41 La formation 42 43 44 45 46 47 48 49 Transfert et partage des informations Learning Processes Composite 50 51 52 53 54 55 56 57 Leadership qui renforce

74.28 9.80 5.72 9.06 9.80 9.52 7.03 50.72 6.30 4.60 7.12 5.84 6.61 5.75 7.92 6.52 50.66 61.01 8.93 9.40 10.06 10.06 10.06 8.45 8.33 8.63 73.93

45.76

63.64

72.91

85.05

97.16

14.28

35.70

49.98

63.67

97.58

14.32 37.48

38.04 51.57

52.79 62.87

62.64 68.87

87.62 82.20

45.00

67.50

75.00

82.50

92.50

267

Appendix G: Tables of Survey Results by Department and by Level of Hierarchy Operations Department AVERAGES

MIN

Q1

Median

Q3

MAX

Question/Section 3 4 5 6 7 Liberté d'expression et droit à l'erreur 8 9 10 11 Appréciation des différences 12 13 14 15 Ouverture à l'égard des idées nouvelles 16 17 18 19 20 Temps pour la réflexion Learning Environment composite 21 22 23 24 Expérimentation 25 26 27 28 29 30 Information collection 31 32 33

19.43 13.44 16.30 18.01 12.86 80.04 21.78 19.65 19.99 21.78 83.21 24.64 20.36 24.24 17.85 87.09 9.72 14.87 9.71 8.87 10.01 53.17 75.88 17.31 16.78 13.21 16.54 63.84 7.85 12.61 9.28 12.85 10.47 11.90 64.97 15.73 17.16 14.87

65.74

74.35

80.03

87.89

91.48

60.69

77.66

87.48

89.26

96.42

67.85

81.82

92.84

95.53

100.00

31.44 63.47

40.76 71.90

55.77 77.52

62.21 79.49

82.88 84.31

32.13

53.85

65.12

77.65

89.27

40.46

49.98

66.64

77.95

88.06

268

34 35 Analyses 36 37 38 39 40 41 La formation 42 43 44 45 46 47 48 49 Transfert et partage des informations Learning Processes Composite

15.73 16.87 80.34 7.85 3.57 7.85 10.47 9.04 5.95 44.74 4.65 3.58 5.01 4.30 5.01 3.58 8.06 6.27 40.45 58.87

62.92

71.50

81.50

90.05

97.16

14.28

41.06

49.98

51.77

57.12

14.32 42.52

25.51 50.62

42.96 61.58

53.25 63.74

64.44 72.58

Commercial/Marketplace Department AVERAGES MIN

Q1

Median

Q3

MAX

Question/Section 3 4 5 6 7 Liberté d'expression et droit à l'erreur 8 9 10 11 Appréciation des différences 12 13 14 15 Ouverture à l'égard des idées nouvelles 16 17 18 19 20

18.42 9.22 15.88 18.73 13.66 75.91 19.04 14.28 11.11 17.85 62.28 21.82 19.04 21.03 15.47 77.36 10.80 15.88 10.80 10.80 11.76

57.20

65.74

74.32

85.72

97.16

49.98

53.55

57.14

64.26

96.42

46.41

71.42

78.55

82.14

96.42

269

Temps pour la réflexion Learning Environment composite 21 22 23 24 Expérimentation 25 26 27 28 29 30 Information collection 31 32 33 34 35 Analyses 36 37 38 39 40 41 La formation 42 43 44 45 46 47 48 49 Transfert et partage des informations Learning Processes Composite 50 51 52 53 54 55 56 57 Leadership qui renforce

60.05 68.90 17.45 19.44 14.68 15.87 67.44 11.64 13.49 13.75 13.75 12.43 8.73 73.78 13.35 14.93 13.66 16.20 14.93 73.07 11.37 5.82 9.82 11.90 9.78 8.20 55.80 6.36 5.77 9.14 5.77 9.34 7.96 8.74 6.36 59.44 65.91 8.33 9.72 10.56 10.56 10.28 7.78 8.06 6.67 71.94

25.74 58.40

57.20 60.37

62.92 64.84

65.78 75.38

77.18 82.52

46.41

60.69

64.26

74.99

89.28

61.88

69.02

73.78

76.16

90.44

54.34

62.92

71.50

80.08

97.16

26.18

47.60

54.74

69.02

90.44

37.59 47.18

46.51 63.29

62.67 66.61

71.54 71.31

75.15 75.84

50.00

70.00

75.00

75.00

87.50

270

Human Resources Department AVERAGES MIN

Q1

Median

Q3

MAX

Question/Section 3 4 5 6 7 Liberté d'expression et droit à l'erreur 8 9 10 11 Appréciation des différences 12 13 14 15 Ouverture à l'égard des idées nouvelles 16 17 18 19 20 Temps pour la réflexion Learning Environment composite 21 22 23 24 Expérimentation 25 26 27 28 29 30 Information collection 31 32 33 34 35 Analyses 36 37

19.43 12.58 17.73 20.00 16.58 86.32 22.14 19.99 16.19 19.28 77.60 24.28 20.00 19.99 17.14 81.41 17.15 16.01 16.01 16.01 17.16 82.34 81.92 17.85 15.00 16.42 17.85 67.12 8.57 13.80 10.95 12.38 9.04 12.38 67.12 15.44 16.58 17.72 14.87 16.58 81.20 14.76 10.95

74.30

82.92

85.76

91.46

97.16

57.12

66.68

85.68

85.7

92.84

64.26

67.84

85.69

89.28

100

65.76 68.23

80.06 81.25

80.08 83.59

88.64 84.56

97.16 91.98

35.70

60.69

67.83

85.68

85.71

40.46

61.88

71.4

71.4

90.44

65.78

71.5

82.92

91.48

94.32

271

38 39 40 41 La formation 42 43 44 45 46 47 48 49 Transfert et partage des informations Learning Processes Composite 50 51 52 53 54 55 56 57 Leadership qui renforce

14.76 12.85 13.33 12.38 79.02 9.30 7.16 6.09 7.52 7.52 7.88 10.37 11.08 66.91 72.27 10.00 9.50 9.00 10.50 10.00 10.50 11.00 10.00 80.50

45.22

73.78

88.06

90.44

97.58

28.64 44.30

55.46 75.42

80.55 78.51

82.28 80.93

87.62 82.20

65.00

77.5

82.5

85

92.5

E-commerce Department AVERAGES MIN

Q1

Median

Q3

MAX

Question/Section 3 4 5 6 7 Liberté d'expression et droit à l'erreur 8 9 10 11 Appréciation des différences 12 13 14 15 Ouverture à l'égard des idées nouvelles 16 17

17.15 10.19 15.19 15.55 14.12 72.20 18.07 20.08 13.83 18.97 70.96 22.99 19.86 17.63 16.29 76.77 7.33 11.44

51.48

59.35

70.07

84.33

94.32

39.27

63.37

76.76

78.54

92.84

53.55

71.41

78.55

83.90

100.00

272

18 19 20 Temps pour la réflexion Learning Environment composite 21 22 23 24 Expérimentation 25 26 27 28 29 30 Information collection 31 32 33 34 35 Analyses 36 37 38 39 40 41 La formation 42 43 44 45 46 47 48 49 Transfert et partage des informations Learning Processes Composite 50 51 52 53 54 55

6.20 6.26 6.08 37.31 64.31 17.18 17.85 11.16 11.31 57.49 9.52 13.24 10.26 11.75 11.61 11.90 68.28 12.87 13.05 13.94 13.59 15.01 68.45 8.03 5.06 7.29 6.99 8.18 4.91 40.46 5.59 3.24 7.49 6.04 5.82 5.37 6.27 5.03 44.86 55.91 7.97 8.59 9.84 8.75 9.06 7.50

14.30 43.94

30.75 56.71

34.32 64.39

42.90 72.08

74.36 85.38

28.56

45.52

58.91

68.72

82.13

38.08

57.72

72.59

76.83

90.44

45.76

61.45

67.21

77.94

91.48

14.28

25.59

36.89

60.10

69.02

17.90 37.48

35.80 48.64

44.75 57.07

55.94 64.54

62.65 70.91

273

56 57 Leadership qui renforce

7.66 8.28 67.66

45.00

62.50

68.75

75.00

85.00

Executive Committee Averages MIN

Q1

Median

Q3

MAX

Question/Section 3.00 4.00 5.00 6.00 7.00 Liberté d'expression et droit à l'erreur 8.00 9.00 10.00 11.00 Appréciation des différences 12.00 13.00 14.00 15.00 Ouverture à l'égard des idées nouvelles 16.00 17.00 18.00 19.00 20.00 Temps pour la réflexion Learning Environment composite 21.00 22.00 23.00 24.00 Expérimentation 25.00 26.00 27.00 28.00 29.00 30.00 Information collection 31.00 32.00 33.00

18.58 7.15 15.02 17.16 15.02 72.92 13.39 17.85 14.28 16.96 62.48 18.75 19.64 21.42 13.39

57.20

65.78

75.77

82.91

82.92

49.98

58.02

62.48

66.94

74.98

73.19 8.58 14.30 7.87 8.58 10.01 49.34 64.48 18.74 19.64 12.50 16.07 66.94 8.33 11.31 10.71 11.31 8.93 7.14 57.72 13.58 15.02 17.15

46.41

65.15

78.55

86.59

89.26

42.90 58.95

45.05 60.02

48.62 63.32

52.91 67.78

57.20 72.34

60.69

63.37

64.26

67.83

78.55

30.94

54.15

63.07

66.64

73.78

274

34.00 35.00 Analyses 36.00 37.00 38.00 39.00 40.00 41.00 La formation 42.00 43.00 44.00 45.00 46.00 47.00 48.00 49.00 Transfert et partage des informations Learning Processes Composite 50.00 51.00 52.00 53.00 54.00 55.00 56.00 57.00 Leadership qui renforce

14.30 17.16 77.20 13.09 8.33 11.90 12.50 11.90 12.50 67.24 7.60 7.61 8.94 6.71 7.61 7.16 9.40 8.06 63.08 66.43 7.50 7.50 10.00 8.13 10.63 8.75 7.50 8.13 68.13

65.78

72.19

78.61

83.62

85.78

54.74

61.88

66.64

72.00

80.92

57.25 63.25

58.62 63.28

59.97 65.59

64.43 68.74

75.15 71.31

62.50

66.25

67.50

69.38

75.00

Managers Average

MIN

Q1

Median

Q3

MAX

Question/Section 3.00 4.00 5.00 6.00 7.00 Liberté d'expression et droit à l'erreur 8.00 9.00 10.00 11.00 Appréciation des différences 12.00

19.15 10.30 16.58 17.43 15.72 79.18 22.50 19.65 16.42 20.35 78.92 23.93

60.06

69.34

84.32

87.89

94.30

49.98

70.60

82.13

88.37

96.42

275

13.00 14.00 15.00 Ouverture à l'égard des idées nouvelles 16.00 17.00 18.00 19.00 20.00 Temps pour la réflexion Learning Environment composite 21.00 22.00 23.00 24.00 Expérimentation 25.00 26.00 27.00 28.00 29.00 30.00 Information collection 31.00 32.00 33.00 34.00 35.00 Analyses 36.00 37.00 38.00 39.00 40.00 41.00 La formation 42.00 43.00 44.00 45.00 46.00 47.00 48.00 49.00

20.00 20.35 17.50 81.77 10.87 15.16 9.44 8.58 10.01 54.05 73.48 18.21 17.49 14.28 14.64 64.62 9.76 13.57 10.95 12.14 11.43 13.09 70.93 14.01 15.73 14.87 15.16 16.58 76.34 10.71 4.76 8.09 8.81 10.23 7.14 49.74 6.62 4.48 8.41 5.19 7.52 6.62 7.52 6.26

57.12

74.97

82.14

91.05

100.00

28.60 59.30

36.47 64.87

58.63 73.06

65.78 83.32

80.06 85.38

28.56

52.66

73.19

77.65

89.27

38.08

64.26

77.40

83.90

88.06

51.48

69.36

75.79

85.05

97.16

26.18

36.89

47.60

56.53

90.44

276

Transfert et partage des informations Learning Processes Composite 50.00 51.00 52.00 53.00 54.00 55.00 56.00 57.00 Leadership qui renforce

52.62 62.85 10.00 11.00 11.25 11.25 10.50 9.50 9.25 9.00 81.75

23.27 42.52

43.41 58.13

54.58 64.94

64.00 69.82

80.55 82.20

70.00

76.88

83.75

85.00

90.00

Employees AVERAGES MIN

Q1

Median

Q3

MAX

Question 3.00 4.00 5.00 6.00 7.00 Liberté d'expression et droit à l'erreur 8.00 9.00 10.00 11.00 Appréciation des différences 12.00 13.00 14.00 15.00 Ouverture à l'égard des idées nouvelles 16.00 17.00 18.00 19.00 20.00 Temps pour la réflexion Learning Environment composite 21.00 22.00 23.00 24.00 Expérimentation 25.00 26.00 27.00

17.97 11.85 15.93 17.56 13.48 76.78 19.25 18.49 15.13 19.64 72.51 23.72 19.77 20.39 16.70 80.58 9.91 13.58 9.26 9.60 9.70 52.05 70.48 16.64 17.47 13.01 14.15 61.27 9.01 12.75 10.71

51.48

67.90

77.25

89.34

97.16

39.27

59.80

74.98

85.68

96.42

53.55

71.41

78.56

92.84

100.00

14.30 43.94

33.50 62.69

55.77 73.59

66.48 78.45

97.16 91.98

35.70

51.77

62.48

68.73

89.28

277

28.00 29.00 30.00 Information collection 31.00 32.00 33.00 34.00 35.00 Analyses 36.00 37.00 38.00 39.00 40.00 41.00 La formation 42.00 43.00 44.00 45.00 46.00 47.00 48.00 49.00 Transfert et partage des informations Learning Processes Composite 50.00 51.00 52.00 53.00 54.00 55.00 56.00 57.00 Leadership qui renforce

12.41 10.88 10.88 66.64 14.09 14.71 14.19 14.81 15.32 73.12 9.01 5.70 9.10 9.78 8.93 6.21 48.71 6.01 4.22 6.39 5.94 6.14 5.24 7.86 6.39 48.19 59.58 8.75 9.11 9.64 9.91 9.82 8.04 8.13 8.57 71.96

33.32

57.72

70.21

76.16

90.44

45.76

62.92

71.50

83.64

97.16

14.28

35.11

48.79

60.10

97.58

14.32 37.48

35.80 49.16

46.53 61.09

58.61 65.67

87.62 80.93

45.00

66.88

75.00

78.13

92.50

278

Appendix H: Five Number Summaries, Box Plots, and Histograms of Survey Results Five Number Summaries of Survey Scores for Building Blocks (with Box Plots): Building Block 1: Learning Environment Psych Safety

Appreciation of Differences

Openness

Time for Reflexion

Learn Environment Composite

Min

51.5

39.3

46.4

14.3

43.9

Q1-Min

17.1

21.4

25.0

20.0

18.9

Median-Q1

10.1

14.3

8.7

17.2

9.3

Q3-Median

9.2

10.7

12.7

14.3

6.8

Max-Q3

9.3

10.7

7.2

31.4

13.1

IQR

19.3

25.0

21.4

31.5

16.1

Range

45.7

57.1

53.6

82.9

48.1

Building Block 1 120.0 100.0 80.0 60.0 40.0 20.0 0.0 Psych Safety

Apprc Differences

Openness

Time for ref

Learn Env Comp

Building Block 2: Concrete Learning Processes and Practices; and Building Block 3: Leadership that Reinforces Learning Experimentation

Information Gathering

Analysis

Education Training

Information Transfer

Learning Process Composite

Leadership that reinforces

Min

28.6

30.6

45.8

14.3

14.3

37.5

45

Q1-Min

25.0

29.5

17.8

21.4

23.7

14.1

22.5

Median-Q1

10.7

11.3

9.3

14.3

14.8

11.3

7.5

Q3-Median

10.7

4.8

12.2

13.7

9.8

6

7.5

Max-Q3

14.3

14.2

12.1

33.9

25

13.3

10

IQR

21.4

16.1

21.5

28

24.6

17.3

15

279

Building Blocks 2 and 3 120 100 80 60 40 20 0

Five Number Summaries of Combined Survey Item Scores (with Histograms): Psychological Safety Min

51.48

Max

97.16

Avg

76.99

Sigma

12.87

Kurosis

-0.93252

Skew

-0.29488

Psychological Safety 14 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

280

Appreciation of Differences Min

39.27

Max

96.42

Avg

73.08

Sigma

14.98

Kurosis

-0.90809

Skew

-0.27589

Appreciation of Differences 12 10 8 6 4 2 0 1

2

3

4

5 Scores

6

7

8

9

10

Calculated

281

Openness to New Ideas Min

46.41

Max

100

Avg

80.16

Sigma

13.30

Kurosis

-0.19054

Skew

-0.48522

Openness to New Ideas 14 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

282

Time for Reflection Min

14.30

Max

97.16

Avg

52.27

Sigma

20.71

Kurosis

-0.74422

Skew

0.085022

Time for Reflection 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

283

Learning Environment Composite Min

43.94

Max

91.98

Avg

70.62

Sigma

10.87

Kurosis

-0.45698

Skew

-0.33254

Learning Environment Composite 18 16 14 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

284

Experimentation Min

28.56

Max

89.28

Avg

62.61

Sigma

16.28

Kurosis

-0.59116

Skew

-0.25664

Experimentation 14 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

285

Information Collection Min

30.94

Max

90.44

Avg

66.81

Sigma

16.60

Kurosis

-0.5271

Skew

-0.57462

Information Collection 16 14 12 10 8 6 4 2 0 10

20

30

40 Scores

50

60

70

80

90

Calculated

286

Analysis Min

45.76

Max

97.16

Avg

74.28

Sigma

13.26

Kurosis

-0.73

Skew

-0.09

Analysis 14 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

287

Education and Training Min

14.28

Max

97.58

Avg

50.72

Sigma

21.55

Kurosis

-0.38

Skew

0.22

Education and Training 10 9 8 7 6 5 4 3 2 1 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

288

Information Transfer Min

14.32

Max

87.62

Avg

50.66

Sigma

18.33

Kurosis

-0.62

Skew

0.00

Information Transfer 10 9 8 7 6 5 4 3 2 1 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

289

Learning Process Composite

Learning Process Composite 18 16 14 12 10 8 6 4 2 0 10

20

30

40

50 Scores

60

70

80

90

100

Calculated

Leadership Min

45.00

Max

92.50

Avg

73.93

Sigma

11.53

Kurosis

0.15

Skew

-0.64

18

Leadership That Reinforces

16 14 12 10

Scores

8

Calculated

6 4 2 0 40

50

60

70

80

90

100

290

Appendix I: Causal Mapping of Individual Positions High performance of brand managers: High Performance of a Brand Manager: successful positioning, visibility, and development of the brands. Factor A: Client's e-commerce turnover Cause: Brand Manager's entrepreneuria l spirit and business sense Knowledge required: technical knowledge of ecommerce, understanding the client's products, final consumer behavior, how to anticipate and imagine the future of the client's ebusiness

Factor B: Good relationship with client Cause: coaching and modelling , practice and feedback from more experienced colleagues/m anagers Knowledge required: how to work in a dynamic, fluid environment with loosely defined processes; ability to anticipate and deal with technical problems

Example 1: one brand manager said he came to the job with some technical skills using the Magento platform but Example: A brand he mainly taught himself manager constantly how to use it. He learned asks the E-store the strategic aspects of managers under her his activity on the job. responsibility questions about the products, sales, visual display, and media investment of the e-stores they manage.

Factor C: Team development Cause: productive team dynamic Knowledge required: people skills, ability to work in a team, communicatio n and listening skills Example (given by e-commerce manager): observed behavior of brand managers who "move around", who are "fed by the teams" , and who "feed the teams:"

Example 2: a brand manager describes complex problem with a brand that does not provide reliable information about its stock and its prices. There is no formal process to deal with this. He has weekly meetings with the e-commerce manager. He must write reports.

Factor D: effective analysis and problem solving Cause: ability to carry out analyses, draw the right conclusions, think creatively

Knowledge required: global understanding of a brands "ecosystem", ability to test markets, performance before, and after quantitative analysis

Example of a problem: a brand recently acquired by a Turkish company was not providing basic information about the products - they had to make very rough estimates of what was stocked in the warehouses. Sales weren't covering costs and their interlocutors were not effective. The company may have to impose punitivite measures on the client.

The brand manager position is one of the most important positions in the company because brand managers have the highest level of direct responsibility for the brands, and hence towards the clients. Three brand managers report to the e-commerce manager. Each brand manager deals with six or seven sites, and manages three pairs of e-store managers who each directly oversee two sites

291

High performance of e-store managers:

E-store success/high performance Factor A: selling the right product at the right time Causes: effective coordination across departments, effective traffic analysis, timely promotions, attitude of client about its own brands

Knowledge required: technical skills for traffic analysis, understanding of the client and business cycles (good timing), creative design of the site. Bad Example 1: an external service provider was not ready on time negativley impacted a sales promotion. Bad example 2: recently during a migration of a client site (to a new platform or server) there were unanticipated technical problems which affected the site. The client was not informed and is not happy. The estore manager now has to find some concession to make the client happy.

Factor B: meeting the "selly-by date": i.e. the site is ready for the client on schedule Cause: feasible "sell by dates", effective coordination and communicatoin across departments and teams.

Example 1: an estore manager says success is when they "explode" sales records for the client's site.

Metaphor/Expression: one employee speaks of working "free style" (refers to cutting corners, thinking creatively under pressure). Example 2: a web designer says: "we always make compromises, we bargain". She states that a special position has been created for this - to deal with "emergencies" Example: 3 one employee says it's not just a question of discounts but of the confidence the client has in its own brand the most gratifying work is with this type of client.

Knowledge required: how to prioritize, how to recognize which technical problems can be resolved after the site is operational.

Example: an employee states that when the sales team has promised an unfeasible delay, delays accumulate and/or the site is delivered with known bugs that have to be resolved after the site goes live.

Example 4: e-store managers work in pairs with the senior manager dealing with business analysis and the junior mangaer dealing with techncial and operational issues.

The above diagram continues with many of the organizational success factors and themes which have been evoked in other tables but drills down more deeply into the daily operations of the firm. The e-store managers, who work in pairs and report to the brand managers, deal most directly with the sites themselves and hence with the technical and creative employees who must carry out the

292

work to make, modify, and maintain the sites. The diagram includes a number of examples, particularly for Factor A (selling the right product at the right time). The examples paint a picture of the everyday experiences and concerns of the e-store managers who must juggle the pressure to generate sales, with the quality of the e-commerce sites, and with the workloads of the various employees who must cooperate in order to carry out work effectively. Furthermore, an e-store manager is a key position when it comes to the flow of information because he or she is often the first to be informed of a technical problem; he or she would thus need to make constant judgments about the gravity of technical problems, and when and how to inform brand managers and/or the clients about such problems. This would be a tacit skill developed on the job through experience and feedback from colleagues and brand managers. The first bad example of Factor A illustrates a situation when the e-store manager is relying on an external provider that isn’t ready on time, and the second bad example illustrates a situation in which the e-store manager is unable to anticipate technical problems related to a site migration and hence cannot warn the client in advance. While the e-store manager must inform the client about the problems in both circumstances, he or she has only limited influence over these matters, and in the second example, presumably the e-store manager was not in possession of the relevant information until after the problems on the client’s site had already started. The other five examples (including the example of the metaphor/expression) provide further illustrations of the preoccupations, concerns and of e-store managers. The expression “free style”, which came from one of the group interviews, suggests that the pressures on the e-store managers cause them to look for ways to cut corners—this example is highly related to the example of Factor C (meeting the “sellby date”) when a site might be delivered with known bugs and the team has to fix the bugs later. In such cases, decisions have to be made about what bugs are acceptable to get the site up and running, so it can function while the team is working behind the scenes to fix the glitches. This may be an example of a tacit skill linked to the corporate culture of the enterprise – i.e. to the assumption (or contractual obligation) that it is better in certain cases to deliver a site with known problems on

293

time than to deliver a site significantly beyond the deadline where all the known problems have been dealt with. Example 2 of Factor A is also noteworthy because it demonstrates that the necessity to make compromises and “bargain” with the client after the service have been sold is so embedded in the company’s way of doing things that they have created a position especially for this.

294

High performance for other positions: digital media manager, customer service manager, head of IT architecture:

High performance: digital media manager Factor: implementaton of new traffic service programs Cause: sucessful relationships built with partners like social media networks Knowledge required: ecommerce knowledge and entrepreneurisal business skills - understanding a P&L statement, etc.

Example 1: According the to ecommerce manager, the digital marketing manager acts almost like an independent entrepreneur starting a new business.

Example 2: developing a successful new strategy for a brand involves setting up new projects and partnerships. Much of this deals with obtaining the rights to exploit user-generated content.

High Peformance: customer service manager

Factor: Number of calls processed and quality of serv ice by customer

service team. Causes: training, team cohesion, smart scheduling

Knowledge required: knowledge of client brands and processes, understanding of losgisitics, technical ability with software, ability to train and motivate service reps.

Example: current manager reduced the number of brands handled by each service rep from 7 to 4. Extensively trains each individual new hire, motivates staff by honoring preferences, (whenever possible), to represent specific brands. A new hire recently requested to work with a particularly difficult brand and after 2 weeks of training was given the brand she wanted.

High performance: head of IT architecture Factor : successful realization of projects, improvements, updates

Cause: successful management of the operations and the people you depend on.

Knowledge required: characteristics of brands, key marketing and sales dates, technical constraints; people and project management skills Example: when you make infrastructure changes , you have to let the client approve of the dates when the site will be down.

Example of a problem: one of their principal clients was recently promised amazing ROI if they used TV advertising as a means to boost e-commerce traffic but no precise analysis was conducted to justify the forecasts. The client communicated the ROI forecasts to its holding company so now the pressure is on to produce results. The actual cost of TV advertising is turning out to be much too high to realistically produce the expected ROI. The employee has to deal with the varous stakeholders to resolve this complciated issue.

The above diagram continues with high performance factors but deals with three very different positions which appear to have little in common: the digital media manager, the customer service

295

manager, and the head of IT architecture. The first two positions are part of the e-commerce department and therefore report to the e-commerce manager, whereas the third position is part of the operations department. The digital media manager position is a relatively new position and has a strong business development dimension to it which requires the person to have an entrepreneurial mindset and a high degree of autonomy; the digital media manager interfaces with other employees and departments at various levels according to the specific partnerships and relationships she is trying to build. Customer service, on the other hand, is a highly repetitive activity which requires tight coordination among members of the team but which interfaces little with the other departments. Meanwhile, the head of IT architecture must work with virtually all the different departments and cooperate in different ways for the realization of specific project and improvements. The example of a problem given by the digital media manager of a problem was elicited during the fifth group interview; it demonstrates a high degree of intrinsic uncertainty linked to the fact that it is a relatively new position and a new activity for the company to be venturing into. The entrepreneurial skills necessary to face this problem would include the ability to face complexity, deal with ambiguity, talk to multiple stakeholders, and find creative solutions. Such skills can be defined as soft skills according to Matteson et al. (2016) and could be considered tacit because they are embedded in practice. The example given by the customer service manager during Individual Interview 7 suggests people management and training skills are most relevant for this position; and it suggests that his approach for this involves a significant degree of feedback and one-on-one coaching. This would be the tacit dimension to an activity (customer service), which is mainly characterized by explicit knowledge, and repetitive tasks. Finally, the example given by the head of IT architecture, illustrates the recurrent need to keep the client in the loop and communicate effectively when carrying out work affecting clients’ e-commerce sites. The head of IT architecture stated during the second group interview that

296

he had been promoted a year ago from web developer to his current position. He has therefore gone from a role where he was primarily executing tasks to a much more complex management position requiring more soft skills and tacit skills. High performance for additional positions: logistics manager, traffic manager, and antifraud officer

High performance: logistics Factor : smooth and timely deliveries, filling customers' orders, good service rate

Cause: technical systems that function well, effective communication Knowledge required: internal understanding of who does what, business analysis, problem identification, how to make recommendations Example (of a problem): recently an employee discovered a technical integration problem which prevented orders from being processed properly. Some were stuck on standby. In other instances, orders were processed but clients were not being charged for the work. She had to peform a cost analysis of the problem and communicate with the various parties to resolve it.

High performance: traffic

High performance: Antifraud

Factor : turnover, number of visits on a site(s) , acquiring new customers for the clients' sites

Factor : preventing fradulent transactions

Cause : convincing the client and colleagues what needs to be done, finding effective ways to boost traffic

Cause : accuracy and frequency in identifying attmpts to make fraudulent transactions

Knowledge required: problem identification, a global understanding of how things work, knowing what effectively boosts traffic Metaphor: it's like understanding how a car motor works. You have to be willing to open up the hood. Example: there are 3 traffic managers who share 11 brands - apparently they each specialize. One works on referencing and display, another on advertising and publicity, etc.

Knowledge required: ability to recognize patterns and type s of fraudulent buyer behavior Examples: with one specific brand, virtually all fraud occurs in one city (Marseille) and typically involves small transactions. With another brand it might be large transactions. Certain delivery modes might be suspect as well as certain types of email addresses.

Example: in the words of one logistics manager, the key to success is having clear, precise, and feasible objectives.

The above diagram maps out additional positions in the company. The logistics manager and the traffic manager both work in the operations department whereas the antifraud officer works in the finance department. The organizational success factor mapped out in this table continues with high

297

performance. The high performance factors for traffic are essentially the same as success factors seen previously: turnover, number of site visits, and acquiring new customers. The logistics and antifraud factors, however, are very specific to their respective activities. The second example given by the logistics manager is actually her own version of a success factor; however, I concluded after listening to two different logistics managers (in Group Interview 2 and 5, respectively), that the real causes for high performance and success in this position were the ones listed in the factor box. Nevertheless, it is instructive to provide the point of view of one of the employees who understandably feels her success can’t be assessed without clear objectives. Meanwhile, the example of the problem given for this activity underscores the importance of vigilance and problem identification and gaining experience dealing with such problems – hence tacit skills. The metaphor provided by the head of traffic in second group interview (opening up the hood of a car to understand how the motor works), illustrates the importance of understanding how the different departments and positions at the company work together in order to successfully reach business objectives such as turnover, site visits, and customer acquisition. As noted in Chapter 3, Ambrosini and Bowman (2001), consider metaphors to be a way of talking about tacit skills that can’t be described directly. Here the head of traffic expresses an idea that is recurrent throughout many of the interviews. It is a simple idea but one that is difficult to explain as a specific skill or set of skills; acquiring an understanding of how different departments and dimensions of e-commerce are interrelated, and leveraging that understanding to accomplish tasks as efficiently as possible can only be learned on the job. For example, a traffic employee could learn about the effects of various changes on an e-commerce website and/or of SEO actions taken by directly observing the impact on the traffic to that site; furthermore, he could learn by exchanging with other employees, and by reflecting on the processes involved. The ability to make connections across the spectrum of employee actions and initiatives, SEO, web design, online consumer behavior, site visits and sales is a key tacit skill undergirding many of the other skills needed to succeed in this company.

298

Finally, the example given by the anti-fraud officer, also points to the importance of experience, observation, and an understanding of how the pieces fit together – except, here the skills are more specific to anti-fraud as opposed to the broader activity e-commerce. Patterns in fraudulent transactions can be observed, but different patterns are specific to different brands, so there appear to be few if any underlying principles which make the skills and knowledge involved more easily transferrable. Hence within this specific activity, tacit skills play an important role.

299

Appendix J: KIP Framework Applied to Specific Company Positions Activity

Whatis high performance?

After sales service/ Customer service

Quantitative: number of calls and e-mails sent. Qualitative: good understanding of the brands, ability to talk to customers and resolve problems and potential conflicts.

Activity

What is high performance?

Junior e-store manager

Supporting the senior e-store manager for successful sales and promotions, successful stocking and delivery of products, successful delivery of sites, components of sites, or site modifications on time.

E-store manager (senior)

Selling the right product at the right time, breaking sales records, delivering the site or modification to the client on time.

How is high performance achieved? Combination of clear written processes, and procedures, solid stable organization and routines with team building, guidance, and support.

What needs to be known? Effective and efficient communication skills (written and oral): ability to answer calls, talk to clients, respond to email inquiries rapidly and accurately. Knowledge of brands: their specific characteristics, processes, issues affecting delivery, etc.

How is high performance achieved? Routines and processes for the creation and maintenance of e-commerce sites. Clear knowledge of stocks and delivery times. Effective analysis of traffic and SEO.

What needs to be known?

Targeted promotional initiatives such as newsletters. Effective technical and operational knowledge and support which is built on routines and processes, availability of information

Understanding of the product and a business sense for selling products. A sense of good timing for sales and promotions, which products to feature and display on line. Sufficient technical

Technical and operational knowledge of ecommerce platforms, systems, and site maintenance. How to prioritize problems and find solutions. Operational knowledge of how to work with external service providers How to work under pressure: “free style”.

How does knowing take place? Introductory training, access to stored information through manuals and diagrams. Quite intensive onthe-job training with constant feedback with managers and colleagues.

How is knowing applied?

KIP type

Ideal work design will involve replication and reuse of routines, storage and easy access to work practice information, continuous process and outcome based performance feedback.

Enacted information (low tacitness, high learnability) with elements of . Apprenticed Know-how (high tacitness, high learnability) introduced by the Customer Service Manager who stresses oral transmission, coaching, and a team dynamic.

How does knowing take place? Technical knowledge through written processes and procedures and learning from doing. Operational knowledge and skills learned from experience and exchanging with colleagues to understand how different pieces fit together.

How is knowing applied?

KIP type

Ideal work design would be largely based on replication and reuse of routines and work settings for storage and easy access to work practice information. But it would also include the application of experience based skills, apprentice/mentor relationships, and clear outcomes expectations and feedback.

Business sense has to be honed and cultivated through experience and being “open to learning from the market”. Technical knowledge and skills can be transmitted through written

Ideal work design would build on routines and work settings that facilitate storage and access to work practice information while providing opportunities for mentorship and incident –triggered outcomes-based performance

This is a layered position. To the extent that it is provides technical and operational support to the senior e-store manager, it is primarily Enacted Information (low tacitness, high learnability) with some limited aspects of Accumulated Information (low tacitness, low learnability). This is a layered position combining aspects of Enacted Information (low tacitness, high learnability), and Accumulated Information (low tacitness,

300

about products, and logistics.

Activity

What is high performance?

Lead Architect (in charge of IT infrastructure)

Successful realization of projects, improvements, updates

Supply Chain Logistics Manager

Product Manager (Saas)

knowledge to talk to technical people, and understand and anticipate technical constraints. How to communicate effectively with clients. How to work effectively with junior estore manager.

processes and procedures.

How is high performance achieved? Application of extensive and complex expertise through the use of logical and systematic analysis.

What needs to be known?

Smooth and timely deliveries, filling customer’s orders, good service rate.

Application of extensive and complex expertise through the use of logical and systematic analysis.

Client autonomy – the client anticipates and makes good decisions.

Application and adaptation of extensive and complex expertise. Effective internal communication so unrealistic promises aren’t made to client. Must work closely with project managers.

How to monitor stock, negotiate with the transporter, understanding physical and digital flow of goods and info, how to deal with discrepancies Business analysis, diagnostics, and how to make recommendations. Knowledge and skills related to web design and development. Ability to use SaaS. Understanding of the business dimensions of ecommerce.

Key cycles and dates for brands, understanding of technical constraints; how to manage others. Extensive, complicated and detailed information that must be understood in relation to contingent factors for situated analysis.

feedback. Furthermore, there would be systematic but flexible ways to transfer and share expertise.

low learnability). It builds on the former, which corresponds to the more technical (junior) aspects of the position in order to facilitate the latter, the business analysis aspect.

How does knowing take place? Extensive study and prior experience to master specialized expertise, numerous systematic and analytic search processes and problem solving accompanied by continuous learning, adaptations, and synthesis of new information. Similar to the above

How is knowing applied? Ideal work design would encourage the application of specialized expertise, systematic but flexible ways to share and transfer this expertise, opportunities for education and reflection, outcomes and process-based performance feedback. Similar to the above

KIP type

Extensive study and training to master specialized expertise. Learning through doing for SaaS (trial and error, practice). Exchanging with colleagues to learn the business side.

Similar to the above

Accumulated Information (low tacitness, low learnability) with aspects of Apprenticed Know-how (high tacitness, high learnability) when it comes to learning while doing and refining techniques.

Accumulated Information (low tacitness, low learnability)

Accumulated Information (low tacitness, low learnability)

301

Activity An e-commerce position embedded in the client company.

Activity Head of Traffic and CRM

What is high performance? Optimal balance between the client’s ecommerce autonomy and the company’s support and guidance.

What is high performance? Effective, customdesigned commercial offers for clients. Associated KPI’s (turnover, number of visits, acquired customers)

How is high performance achieved? A near perfect sharing of information so that each side can fulfill its role.

How is high performance achieved? Finding effective ways to boost the traffic on the ecommerce sites. Convincing clients and colleagues what needs to be done, making sure they are aware of the results of your work.

What needs to be known? Broad scope of knowledge about the client brand, its corporate culture and practices. Deep understanding of e-commerce, the services provided by the company, and the practices and culture of the company.

What needs to be known? Knowledge and skills required are complex and constantly evolving: how to find new users, digital referencing and display, knowledge of advertising and PR, notions of ROI. Extensive knowledge needed to find the solutions which best fit the problems. Effective skills to communicate with the client. You need a “vision of the market”.

How does knowing take place? Total immersion in the client company to assimilate its culture and practices. Learning by doing (ecommerce platforms). Exchanging and proactively engaging with interlocutors (both client and company).

How does knowing take place? Robust study and technical expertise coupled with on the job learning to apply and refine the knowledge. The metaphor was suggested of opening up the hood of the car to understand how the motor works – but substantial training and knowledge are required to make sense of “the motor”. Feedback with managers, colleagues, and

How is knowing applied?

KIP type

Ideal work design: Accumulated information cut and paste

This position is partly Accumulated Information (low tacitness, low learnability) – the person functions like a consultant, an auditor or analyst working in the client firm. But there is a high degree of adaptability required (hence tacit skills), which suggests the position could be very different depending on the client. There are elements of Apprenticed Know-how, and Talent & Intuitive Know-how.

How is knowing applied?

KIP type

Ideally work would be designed to facilitate the application of the specialized expertise with ways to share and transfer knowledge with colleagues. There should be opportunities for reflection and performancebased feedback. There should also be new challenges and opportunities for self-leadership.

Hybrid between Accumulated Information (low tacitness and low learnability) and Apprenticed Know-how (high tacitness and high learnability) with dimensions of Talent and Intuitive Know-how (high tacitness, low learnability)

302

Brand Manager (e-commerce)

Activity

Successful positioning, visibility and development of the brands.

What is high performance?

Web designer

Successful and timely delivery of e-commerce sites, components of sites, or modifications, to the client. Visual quality and coherence of the site. Client satisfaction.

Digital Media Manager

Business development – in the future,

Anticipating trends and creatively thinking about how the brands can develop, maintaining good relations with the client, developing the teams working with the brands.

A global understanding of the “brand’s ecosystem”, a deep understanding of e-commerce and e-commerce platforms, a strategic business sense of what drives traffic and turnover, how to communicate with the client, people management and team building skills

How is high performance achieved? Alignment of the following items: promises made to the client/expectations of the client; what is technically feasible, given technical and competency constraints; and what is operationally feasible, given workload and time constraints. Making compromises that are satisfactory to the client

What needs to be known?

Entrepreneurial attitude, going to talk to clients,

Knowledge of social media and how they can

Technical knowledge of how to conceive, set-up, maintain, and modify sites and e-shops. People skills when it comes to dealing with clients and colleagues. How to talk to the marketing department, web integrators, and web developers. How to make compromises and handle emergencies (last minute changes, technical glitches).

team members are important factors. A brand manager must have learned and assimilated robust technical and business knowledge through study and past experience. On top of that, brand managers must learn on the job – feedback and mentoring play a significant role.

How does knowing take place? According to one employees, 80% comes of the knowledge required for this position comes from studies and technical training, and 20% is learned on the job through a mentor-like relationship with a peer.

Much of the knowing will be gained through

Ideally work would be designed to facilitate the application of the specialized expertise with ways to share and transfer knowledge with colleagues. But there should also be an emphasis on application of experience based skills and developing those skills through related experienced. Apprentice/mentor relationships should have an important role and constant outcomebased feedback. There should be new challenges and opportunities to develop selfleadership.

How is knowing applied? Ideal work design allows for the application of specialized expertise, systematic but flexible ways to transfer and share that expertise. But it would also leverage some unique talent and abilities, and initiate new challenges. There would be natural tension between selfleadership (for the unique talents) and outcome and process-based performance feedback (based on the client satisfaction). Ideal work design will use specialized

Hybrid between Accumulated Information (low tacitness and low learnability) and Apprenticed Know-how (high tacitness and high learnability) with dimensions of Talent and Intuitive Know-how (high tacitness, low learnability)

KIP type

This position is primarily Accumulated Information (low tacitness, low learnability). But there are limited elements of Talent and Intuitive Knowhow (high tacitness, low learnability) when it comes to the creative conception of the sites.

This is a layered position

303

more quantitative KPIs can be established such as turnover generated thanks to this activity.

Activity Cofounder/Head of Photography

What is high performance? Ability to anticipate change and offer new types of services/products needed by the evolving market

seeking new partnerships.

How is high performance achieved? Innovative ideas, willingness to take risks, ability to reach and convince key decision makers for brands in the sector.

work for the brands; how to develop brands thanks to digital media; knowledge of intellectual property rights; understanding how to follow traffic, estimate costs, ROI, and risk.

experience – i.e. trial and error, experimentation, and risk taking. Experience will create opportunities for reflection and doing things different the next time around. There is also an element of mentorship – person can learn from others in the company who must develop business.

expertise to facilitate systematic but flexible ways to transfer and share experience. It will also use and leverage unique talents and abilities, create new challenges, and encourage selfleadership.

including aspects of Accumulated Information (low tacitness, low learnability) but also stressing Talent and Intuitive Know- how (high tacitness, low learnability)

What needs to be known?

How does knowing take place? Studies and experience are required to gain the base of knowledge in ecommerce. Ideal prior experience in brand management (or something closely related). On the job, knowing takes place through conditions that nurture innovation, creativity, aligning diverse efforts, exploration, and experimentation.

How is knowing applied?

KIP type

Ideal work design leverages unique talent and abilities, initiates new challenges. It encourages selfleadership and social network development.

Talent and Intuitive know-how (high tacitness, low learnability)

Evolving understanding of e-commerce trends in the sector, broad and constantly evolving understanding of how the Internet functions and how to innovate on the Internet, knowledge of the potential and limits of technical innovations, knowledge of brands (their products, needs, and customers),

304