Presented By
O’Reilly + Cloudera
Make Data Work
March 25-28, 2019
San Francisco, CA
Please log in

Recommendation engines and mobile gaming

Bysshe Easton (KIXEYE), Thomas Dobbs (KIXEYE)
11:00am11:40am Wednesday, March 27, 2019
Secondary topics:  Media, Marketing, Advertising
Average rating: ****.
(4.50, 2 ratings)

Who is this presentation for?

  • Analytics managers and data scientists



Prerequisite knowledge

  • An understanding of core data science concepts (useful but not required)

What you'll learn

  • Understand how recommendation engines can improve your business
  • Explore considerations when implementing in a production environment for your online product


Mobile gaming is a $50+ billion industry. Much of the industry’s growth has been fueled by the sale of in-game virtual resources and items to help players progress further or improve their overall gaming experience. One of the biggest concerns for mobile gaming developers is improving their overall monetization without getting in the way of players enjoying the game.

KIXEYE—a developer of complex mobile strategy games—periodically provides its player base with a handful of in-app purchase options that provide different in-game content at different price points and discounts. The problem that companies run into using this model is what in-app purchases should be shown and when in order to maximize the number of in-app purchases. This is made more difficult at KIXEYE due to the massive number of in-app purchases available in the company’s games. So how do you solve this problem?

Bysshe Easton and Thomas Dobbs explain how KIXEYE used hybrid recommendation engine techniques to create personalized in-app purchase recommendations for its customers, resulting in a 20%+ lift in user revenue. Along the way, they cover some parallelization techniques the company used to nearly eliminate scaling issues.

Photo of Bysshe Easton

Bysshe Easton


Bysshe Easton is the director of analytics at KIXEYE, where he brings his experience using a combination of data analysis, economics, intuition, and game design sense to solve monetization and content delivery problems in games. He also manages the insights and implementations of the machine learning offer system for War Commander: Rogue Assault.

Photo of Thomas Dobbs

Thomas Dobbs


Thomas Dobbs is a data science product manager at KIXEYE, where he is responsible for maintaining and building machine learning models from ideation to full implementation. Thomas also wrote the underlying algorithm for KIXEYE’s offer recommendation engine. His experience spans marketing, user acquisition, finance, and product, and he has spent seven years in the gaming industry. He is an MBA candidate at UC Berkeley’s Haas School of Business.