Cohort analysis

A cohort in mobile marketing is a set of users that are grouped together due to a common factor. The common identity for a cohort analysis in mobile marketing is typically users with a shared conversion date or those who installed the app on the same date or within the same date range, or perhaps as a result of a common campaign or source.

Cohorts are used in mobile marketing in order to compare and analyze data of users that share similarities. For example, if you’ve made changes to an app and want to examine the impact on users, analyzing a single cohort relevant to these changes and comparing it to other cohorts can lead to better insights than analyzing the entire user base. In addition, if you’re making several changes or improvements to an app, cohort analysis can segment your users to provide a clear understanding of which changes performed the best.

Cohort analysis is important because users have an engagement lifespan after the time they’ve installed your app. If you’re analyzing the entire user base, the changes you’ve made may not be noticeable, which is why it’s important to segment users into cohorts. For example, if you’ve made a change to your app onboarding, you may want to segment users that have just installed the app this week.

As highlighted in this guide to cohort analysis:

By segmenting your sample customer base into groups with shared attributes or experiences like this, you can collect user feedback and track similar actions made by a collective sample group, then begin to assimilate behavioral patterns. Businesses such as app development agencies will use these patterns to make informed decisions on how to improve revenue and user retention rate post-download.

In short, cohort analysis is used to help app publishers optimize various metrics of an app including retention rate, lifetime value, and other key metrics, all of which help improve the app’s overall ROI. The key point of using a cohort is that instead of analyzing thousands or millions of people, you break the user base down into cohorts that can be analyzed more accurately.

As a leader in mobile marketing analytics, Singular helps marketers perform cohort analysis by taking a group of users with a shared conversion date (usually, users who installed an app on a specific date), and looking at their behavior over time. As discussed in our FAQ on the subject, with a cohort analysis, marketers can see:

• How much revenue did this specific group of users generate within 7 days of converting?
• How many of the users are still active by day 14?

A key goal of cohort analysis is to separate user-engagement metrics from growth metrics. By using the conversion date, this lets marketers measure the behavior of existing users over time without being overshadowed by engagement from brand new users.