Login

Gaining marketing knowledge from social media images

Gaining marketing knowledge from social media images

The flood of images 

The lavish party with friends, the holiday on the beach, the newly acquired smartphone - all of this is captured on camera and shared via social networks. The flood of images is large: the photo-sharing platform Flickr, for example, already includes 8 billion images. Around 3.5 million are added every day. 1.15 billion Facebook users upload an average of 350 million photos per day. The entire photo database of Facebook already counts 250 billion.


Highlights:

  • Brand love study with 148,393 social media photos from 503 Germans and Americans 
  • Social media image study of 41 FMCG brands with 47,988 photos, 57,984 textual postings, online survey with 1,000 people and purchasing data from a panel of 30,000 households 
  • Study on consumer-brand interactions with 941,731 social media images 
  • Study on beauty trends with 16,568 social media photos 
  • Study on confectionery brands with 49,074 social media photos 

 

Social media images as a source of knowledge 

These snapshots do not only provide insight into the lives of users, they also reflect their attitudes and experiences towards brands and products. And they influence a potentially large circle of viewers. Images often have a greater impact than text because they are perceived more subtly and have a stronger influence on the emotions of the viewer. User-generated photos also have a high level of credibility compared to professional photos. Social media images, therefore, represent a rich source of data for market research - which, however, has so far hardly been usable because existing tools for social media analysis focus on textual postings.

Computer vision for gaining marketing knowledge 

Due to the large number of images in social networks, manual evaluation is only possible to a limited extent and automated methods for image analysis are necessary. For this reason, we developed the PictureScan tool, which gains marketing-relevant knowledge from user-generated photos. The content of the image will first be recognized using methods from the field of computer vision. Awareness, popularity, usage situations and interactions of brands, products and consumers can be determined through further analysis. These key figures are evaluated in comparison to the competition and over time. In this way, trends can be uncovered and opportunities and risks for corporate image and sales can be estimated.

Awards:

Innovation Award 2016 of the Professional Association of German Market and Social Researchers (BVM) 

Cooperation Partners: 

University of Augsburg, Prof. Dr. Rainer Lienhart
University of Michigan-Dearborn, Prof. Dr. Aaron Ahuvia 

Publications:

  • Harzig, P., Zecha, D., Lienhart, R., Kaiser, C., & Schallner, R. (2019). Image captioning with clause-focused metrics in a multi-modal setting for marketing. 2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), San Jose, CA, USA, 419-424. 
  • Kaiser, C.; Schallner, R.; Manewitsch, V. (2019). Revealing Consumer-Brand-Interactions from Social Media Pictures – A Case Study from the Fast-Moving Consumer Goods Industry, Proceedings of the 21st General Online Research Conference, Cologne. 
  • Harzig, P., Brehm, S., Lienhart, R., Kaiser, C., & Schallner, R. (2018). Multimodal image captioning for marketing analysis. Proceedings of the IEEE Conference on Multimedia Information Processing and Retrieval, FL, USA. 
  • Kaiser, C., Enzingmueller, L., & Schallner, R. (2018). Analyzing the temporal development of brand-related social media photos. GfK Verein Working Paper Series, 6. 
  • Kaiser, C. (2017). Ein Bild sagt mehr als Tausend Worte - Neues Marketing-Wissen aus Social Media Fotos ziehen. Planung&Analyse, 1, 51-53. 
  • Kaiser, C., & Ahuvia, A. (2017). Are Facebook brand pictures a sign of brand love? – A study on visual eWOM. Proceedings of the 3rd International Colloquium on Corporate Branding, Identity, Image and Reputation, London. 
  • Paolanti, M., Kaiser, C., Schallner, R., Frontoni, E., & Zingaretti, P. (2017). Visual and textual sentiment analysis of brand-related social media pictures using deep convolutional neural networks. In S. Battiato , G. Gallo, R. Schettini & F.  Stanco (Hrsg.), Lecture Notes in Computer Science. Presented at the International Conference on Image Analysis and Processing (S. 402–413). Catania: Springer. 
  • Fabian Buder, Thomas Couronné, Carolin Kaiser (2017). Digitalization and the value of new (meta) data sources for market insights: How increasing options and new decision requirements can impact the market research value chain. Proceedings of the SAMRA Annual Conference, Cape Town. 
  • Carolin Kaiser (2017). Ein Bild sagt mehr als Tausend Worte - Neues Marketing-Wissen aus Social Media Fotos ziehen. Planung&Analyse 1/2017, 51-53. 
  • Carolin Kaiser (2016). Ein Bild sagt mehr als Tausend Worte. Markenartikel 10/2016, 62-63. 
  • Carolin Kaiser, Raimund Wildner (2016): Gaining Marketing-Relevant Knowledge from Social Media Photos - A picture is worth a thousand words. Proceedings of the 2016 ESOMAR Congress, New Orleans. 
  • Carolin Kaiser, Aaron Ahuiva, Philipp Rauschnabel, Matt Wimble (2016): Visual eWOM: Are Facebook Brand Photos a Sign of Brand Love? 2016 AMA Summer Marketing Educators' Conference, Atlanta. 
  • Carolin Kaiser, Lisa Frey, Björn Ivens (2016), What Photos Reveal: Recognizing Consumer Brand Relationships from Social Media Photos, GfK Verein Working Paper Series, No. 2/2016 
  • Kaiser, C. (2014). Soziale Medien als Mittel der Produktgestaltung (Co-Creation). In: König C., Stahl M., Wiegand E. (eds) Soziale Medien (pp. 171–194). Springer VS, Wiesbaden. 
  • Carolin Kaiser, Lisa Frey, Björn Ivens (2014), Characterizing Consumer-Brand-Relationships in Social Media Pictures, Proceedings of the 43rd Annual European Marketing Academy Conference, Valencia. 

Events:

  • IEEE Conference on Multimedia Information Processing and Retrieval, March 2019, San Jose 
  • General Online Research, March 2019, Cologne 
  • IEEE Conference on Multimedia Information Processing and Retrieval, April 2018, Miami 
  • Predictive Analytics World, November 2018, Berlin 
  • South African Marketing Research Association Annual Conference, Oct 2017, Cape Town 
  • International Colloquium on Corporate Branding, Identity, Image and Reputation, Sept 2017, London 
  • International Conference on Image Analysis and Processing, Sept. 2017, Catania 
  • Bayreuth Economic Congress, May 2017, Bayreuth 
  • Photo Industry Association Congress, March 2017, Frankfurt 
  • AMA Summer Marketing Educators' Conference, Aug. 2016, Atlanta 
  • ESOMAR Congress, Sept. 2016, New Orleans 
  • Ludwig Ehrhard Symposium, August 2016, Nuremberg 
  • Research Plus, July 2016, Nuremberg 
  • BVM Congress, April 2016, Berlin 
  • European Marketing Academy Conference, June 2014, Valencia 
  • Social media conference of the Federal Statistical Office, June 2013, Wiesbaden