Higher Retention and LTV Results with Machine Learning

Due to the nature of the mobile game industry, it isn't easy to find users organically. Mobile game owners or publishers spend most of their marketing budgets on ads needed to acquire users until they arrive organically. In other words, mobile game companies spend all their marketing budget on user acquisition.

In the beginning, targeting the right user segment with optimized ads campaigns and acquiring them is enough. After installing the game to the users, it is important to keep them in the game as long as possible (Higher Retention). So, the longer the users stay, the more profit they leave.

In this case, we would like to present our ongoing partnership with a well-known RPG game.

The goal of the collaboration was to increase the Retention of Players. Therefore, the first step was to distinguish between players who will stay in the game less than 7-days and those who will stay longer than 7-days.

Then, AppNava’s 7-Day Retention Model was initiated for Newcomers.

After the player finished the first session, AppNava detected who will leave the game earlier (play less than 7-days). These players have been separated into two different groups for the A/B test.

Let’s dive deeper into the amazing result;

The AI-based retention model learning period ranges from April 1 to April 30, 2021. In total, 1.432.326 Players have been analyzed. At the beginning of the test, the 7-Day Retention Rate was 26%.


Players identified to play the game for less than 7-days have been subdivided into two different control groups to figure out which instruments inspire players to spend a longer time in the game.

We hypothesized that a non-consumable in-game gift would motivate players to play longer.

So, Group B (Green Bar) players have been awarded gifts, while no gifts have been awarded to players from Group A (Red Bar). At the end of the test, the two most crucial in-game metrics, Retention and Live Time Value (LTV) showed observable improvements.


Group A (Red Bar) no gift has been awarded — Group B (Green Bar) has been awarded a gift.

The game's retention rate increased by 13% as a result of accurately predicting players who will leave the game and awarding them with a gift.


The average LTV per Player increased by 55% in Group B by accurately identifying their retention behavior and awarding gifts.

In the end, their monthly revenue increased by $5400.

Retention Rate Increased by 13%

Monthly Revenue Increased by $5400

LTV Increased by 55%

Trusted by leading game studios and publishers

Scale your game with AppNava.