APG is a data-intensive system that automatically creates rich personas representing customer segments by employing web/social media analytics.
Here is an example of an APG persona.
APG uses this analytics data to identify customer behaviors, generates customer segments, and then enriches these customer segments with gender, age, and nationality appropriate names and pictures; customer loyalty rating, customer interests, product interactions, brand sentiment, and segment sizes represented by the personas, … all done in a privacy-preserving process using only aggregated data.
So, APG is an exceptionally data-intensive system!
Here is APG by the numbers! (as of 18 May 2020)
The APG system has more than:
- 1.5K images for personas, copyrights purchased or common use license and meta tagged with gender-age-nationality
- 1M unique names for personas, meta tagged with name-age-nationality
- 178,340 personas generated for current clients over a more than three year period
- 17,465 persona sets (different number of personas, different type of personas (for the month (country), for the month (region), lifetime (country), lifetime (region))
- 621 generations of persona sets
The APG system has identified more than: 8M customer segment sizes (i.e., customer segments represented by personas)
The APG system leverages more than:
- 599K pieces of contents from multiple data sources (currently YouTube, Facebook, Twitter, Instagram, and Google Analytics)
- 28M content comments from multiple data sources (currently YouTube, Facebook, Twitter, and Instagram)
The APG system, for user engagement measures, leverages more than:
- 205M likes from Instagram
- 28M comments from multiple data sources (YouTube, Facebook, Twitter, and Instagram)
The APG system, for appropriate name generation and meta-tagging, leverages more than: 5M publicly available online profiles.
Data-intensive! Data-driven personas!
Jansen, B. J., Salminen, J., and Jung, S.G. (2020) Data-Driven Personas for Enhanced User Understanding: Combining Empathy with Rationality for Better Insights to Analytics. Data and Information Management. 4(1), 1-17. https://content.sciendo.com/view/journals/dim/4/1/article-p1.xml