pLTV Helps Pixio Find and Retain High-Value Users
Campaign Optimization with Machine Learning
Industry: Mobile games
Company Size: 20M+ installs, including top-ranked games for Android/iOS
Solution: Acquire and retain high-value users
Platform Use Case: Campaign optimizationBroadly speaking, optimization is a process used to either maximize or minimize an output value by selecting the right input values. In data science, this… More with machine learning
higher revenue per purchase with app event optimizationApp Event Optimization (AEO) is a type of campaign optimization on Facebook (Meta). The platform looks for users who are mostly likely to complete specific… More for predicted lifetime value custom events, compared to usual campaign
higher incremental return on ad spend with app event optimization for predicted lifetime value custom events, compared to usual campaign
lower cost per install with app event optimization for predicted lifetime value custom events, compared to usual campaign
What Pixio Says About Pecan
Pixio, a mobile gaming studio based in Hong Kong, focuses on developing casual and single-player mobile games that have gained popularity among gamers worldwide. Their hit titles include Tap Tap Trillionaire, Tap Tap Evil Mastermind, Game of Earth, and Summoner’s Greed.
After Summoner’s Greed, Pixio’s flagship title, had been available for three years, the company wanted to broaden its audience and bring in new, high-value players.
Pixio used the Pecan platform to build a highly accurate predicted lifetime value (pLTV) modelIn the context of machine learning, a model is a specific instance or example of an algorithm that has been created based on a particular… More that could predict new players’ lifetime value at Day 30 based on their activity by Day 2. The team then created a Facebook app event optimization campaign using custom predictive events. The campaign specifically sought to connect with players predicted to be in the top 1% of pLTV values.
To validate the incremental value of this approach, Pixio conducted lift tests in partnership with Facebook. They compared the campaign optimized with Pecan’s pLTV predictions to a business-as-usual campaign optimized with purchase events.
Testing revealed that the campaign using Pecan’s pLTV-based predictive event optimization was more successful in attracting and retaining the high-value players who fuel long-term success. The pLTV-informed campaign saw 31% higher revenue per purchase, 18% higher incremental return on ad spend, and 4% lower cost per install.
Overall, adopting a predictive approach allowed Pixio to foresee new players’ long-term value. Predicting LTVLifetime value (LTV) is the total amount of revenue a business can expect to take in from a specific customer over the entire time period… More helped Pixio advertise more efficiently to identify and retain more of the most valuable players, generating greater revenue over time.