Using Big Data in real-life online marketing: Personalitytargeted and tailored advertising on Facebook Matz, S., Chan, A. F., Popov, V., Stillwell, D. & Kosinski, M.
Introduction
Method
Results
Research question
Target group selection
Overall reach: 54,572 users, overall CTR: 2.4% ( = 1,317 clicks) Results of log-linear analysis: 1) Sig. main effect of target group (p < .005, OR = 1.17) 2) Non-sig. interaction effect of target group × advert type (p =.524) 3) Sig. main effect of advert type (p < .005, OR = 1.17
Effectiveness of personality-customised advertising demonstrated in lab studies (1,2,3) However, questionnaire-based nature of personality assessment has largely prohibited its application in real life marketing contexts.
Applying a Big Data approach of predicting personality of Facebook users, we test whether and how personality-based advertisement customisation can be used to increase the effectiveness of online marketing.
Experimental design Advertising campaign: Promotion of Crossword App Examined personality dimension: Extraversion Manipulations of: 1) advertisement style: extraverted vs. introverted 2) Target group: extraverted vs. introverted
Target group selection based on myPersonality database: Contains FB profile information and Big Five results of > 6,000,000 users. Over 36,000,000 User-Like pairs Allows for the extraction of average personality profiles for each “Like” Extraverted Likes
Introverted Likes
Socializing Soulja Boy Partying Parties Akon
The Simlarillion Babylon 5 Computer programming Sonata Arctica Minecraft
Additional criteria: US-based Android users with version 2.3 or higher
Advertisement design Looking for excitement and fun? Play with friends & become a crossword champion!
Hypotheses
H2: Higher click-through-rate (CTR) for conditions in which the advertisement characteristics corresponded to the users Extraversion trait (tailoring effect)
H1: targeting effect
H2: tailoring effect
Non-sig. interaction and sig. main effect of advert type: consumer-tailored vs. product-tailored advertising!? Future research: Under which conditions do consumer- vs. product-tailored approaches yield better results?
2 × 2 between-subjects design
H1: Higher click-through-rate (CTR) for Facebook users low on Extraversion (targeting effect)
Conclusion
Take some time for yourself! Sit down, relax & get immersed in the world of word puzzles.
References Advertisement design based on: (1) General description of Extraversion dimension (2) Semantic analysis of extraverted vs. introverted Facebook users’ status updates (3) Colour preferences of introverts vs. extraverts
www.psychometrics.cam.ac.uk
1 ) Hirsh, J. B., Kang, S. K., & Bodenhausen, G. V. (2012). Personalized Persuasion: Tailoring persuasive Appeals to Recipient’s Personality Traits. Psychological Science, 23(6) 578-581. 2) Sirgy, M.J. (1986). Using self-congruity and ideal congruity to predict purchase motivation. Journal of Business Research, 13(1), 195-206. 3) Moon, Y. (2002). Personalization and Personality: Some Effects of Customizing Message Style Based on Consumer Personality. Journal of Consumer Psychology, 12(4), 313-326.
Sandra Matz (PhD candidate) email:
[email protected] Department of Psychology