Cohort analysis is one of the most effective ways for mobile app developers to analyze user behaviors and improve their appâ€™s engagement, retention and monetization metrics. It lets developers strip away most of the background noise that comes with aggregate reports and capture actionable insights on individual pockets of high-value customers. But for those developers that donâ€™t have teams of data scientists to help implement best practice techniques, the idea of cohort analysis can seem a bit intimidating.
It shouldnâ€™t. The whole point of cohort analysis is to make it quick and easy to understand whatâ€™s really going on within an app and to draw conclusions about the â€śwho,â€ť â€śwhyâ€ť and â€śhowâ€ť of user behaviors. With cohort analysis, developers have the tools that proactively surface data and insights to enable immediate performance improvements.Â
Some definitions of cohort analysis limit it to the study of those groups of users that completed a particular actionâ€”usually an app installâ€”within a specific time frame. Others take a broader approach, defining a cohort as any group of users that share specific metrics, such as an activity, characteristic or feature. Under this latter definition, which happens to be the definition used by Tapjoy, a cohort could include everyone who downloaded an app last month, all of the users in a specific country or region, or everyone who completed the appâ€™s tutorial.
Below, we will explore some of the more basic yet powerful ways of using cohort analysis to better understand and take action based on in-app activity.
Track the impact of app updates
Suppose you recently made changes to your First Time User Experience (FTUE) in order to streamline the introduction and get players into your app faster. Youâ€™d want to know what impact this change had on your engagement and retention. The only way to measure such an impact is to conduct cohort analysis to compare the users who downloaded the app after the change to cohorts who went through the FTUE beforehand. That way you can easily compare, say, the 1-week re-open rate for one cohort versus the other and see whether the change had a positive or negative impact on retention.
Identify your highest value users
Which countries are your biggest spenders and most engaged users coming from? How do iOS users compare to Android users? Smartphone owners vs. tablet owners? You can use cohort analysis to easily compare various segments and identify your highest value users. You can then dig deeper into each of those cohorts to understand how theyâ€™re interacting with your app and how their behaviors differ from other users. This information is invaluable as you build out engagement and monetization campaigns catering to these high-value users.Â
Understand the spending habits of whales vs. dolphins and minnows
When devising campaigns to promote currency sales or IAP offers to paying customers, itâ€™s important to understand the spending habits of your biggest spenders, or â€śwhales,â€ť as it relates to that of your smaller spenders, aka your â€śdolphinsâ€ť or â€śminnows.â€ť With cohort analysis, you can easily gain insights into which items are most popular amongst your various segments of paying customers, at which price points, on which levels, through which channels, and so on, and you can adjust your promotional campaigns accordingly.
Optimize user acquisition campaigns
By comparing a cohort that came from one advertising or referral source to cohorts that came from somewhere else, you can easily see which source drove higher quality users. Youâ€™ll be able to analyze how engaged users from one source are versus the other, what their retention rates and frequency rates are, and what their overall lifetime value is. This information is absolutely vital in optimizing your ad spend and marketing mix to make sure youâ€™re getting a proper return on your user acquisition investment.
Monitor the effect of external factors
Cohort analysis is the best way to understand how external factors affect your appâ€™s retention, engagement or monetization metrics. What happens when a major new competitor enters your space, or the platform makes an important change to its operating system, or the market dynamics shift? Cohort analysis can be used to compare the metrics of users who installed your app before one of these external changes to those who installed it afterwards to give you a clear sense of its potential impact on your app.
Compare beginners to advanced users
Someone who has used your app extensively for months will undoubtedly use it differently than someone who just downloaded it yesterday. Youâ€™ll want to understand these differences so that you can treat these types of users appropriately. How can you engage beginners more deeply and get them to spend money sooner? When is the right time to start monetizing through IAP? Do advanced users open the app as frequently as beginners? Are their sessions longer or shorter? Cohort analysis can help you formulate answers to all of these questions and more.
Conclusion: Donâ€™t be intimidated
Not only do cohort reports make it easy to understand whatâ€™s going on in your app and why users behave the way they do, but they provide an incredibly useful way to study user segments and spot important trends in app usage. There are many good ways to use cohort analysis, and though the list above is by no means exhaustive, it represents some of the more common and effective methods to help get you started. Use these tactics to guide your cohort analysis strategy and youâ€™ll find it easy to optimize the overall health of your appâ€”even if you donâ€™t have a team of data scientists to help you out.