Calculating lifetime value of freemium gamers: calculating retention rates

On April 28, 2011

Matt Tubergen heads Recharge Studios, a wholly owned subsidiary of W3i that invests in the development and marketing/distribution of freemium mobile games.  W3i is a market leader in distributing and monetizing apps with over 500 million apps distributed for W3i clients. Recharge Studios is actively seeking new investment opportunities, if you have a great idea for a game contact us.

In last week’s post we worked towards identifying ways to calculate the Revenue per Daily Active User as part of our Lifetime Value equation. This week, we look at ways to identify another variable in that equation: retention.

What is retention?

Retention is a key metric that speaks to how many users stick with your app day to day after initially being acquired. When attempting to calculate retention rates there are a variety of different ways to drill into the data and group users. Those groups of users, or vintages, may refer to the day they were acquired, the method in which they were acquired (offerwall, organic, advertising etc.)Or the source from which they were acquired (specific publisher source).

Calculating retention rates for mobile apps

Retention is calculated as an average of active users over time. We first have to calculate the daily retention rate on the first day a user is acquired, retention rates are 100%. Let’s say on day 1 100 users are acquired. On day two, 50 of those users are still active, making the retention rate on day two 50%. Now if only 25 of those users return on day three, the third day retention rate for that vintage would be 25%. To calculate the average retention rate for the entire application, we then take an average of all the daily retention rates for the period. . If we want to look at a more formulaic approach we could justify the following:

Retention Rate = Rate X (1 – Attrition Rate), so Day 1 it is (1 – Attrition Rate) and Day 2 it is (x% X (1-Attrition Rate))

For the sake of calculation, we used large, rounded numbers, often times when calculating the retention you’ll work with a larger data set and have a lower average.

Developers can also dive deep to understand specific retention rates by different user groups/vintages. By segregating by vintage, developers can start to identify different trends that will ultimately help make strategic decisions on user acquisition strategies. Consider the following:

User Group1 = Week 1 Users Acquired

Revenues1 = Week 1 Revenues Earned

Rev/User1 = User Group1 Rev/User for Week 1


User Group2 = Week 2 Users Acquired

Revenues1 = Revenues earned by this group in their Week 1

Rev/User1 = User Group1 Rev/User for Week 1

Note on retention rates

This post looks to identify, simply, how to calculate retention but its worth noting that projecting or forecasting retention rates may also come into play. Additionally, there are different ways to look at and interpret the data depending on what kind of information you’re trying to gain from your users.

Next week we’ll bring it all together to look at calculating lifetime value.

Do you have a question about freemium gaming or a topic you’d like us to explore? Let us know in the comments or catch us on twitter @rechargestudios or @w3i.

Freemium Game Blogs are published in partnership with the series on W3i’s corporate blog.