What is a cohort?

In mobile marketing, a cohort is a set of users grouped together because of a common identifier. As long as there is commonality, a cohort can be anything: from users in a certain geographic location to users who installed an app within the same period.

Cohorts are designed to let marketers compare data more easily, using the metrics that matter in a certain context. That being said, there are also many things you can do with cohorts that cement cohort analysis as an essential part of mobile marketing.

Once a user has installed an app, they have an engagement lifespan, which expresses trends you can manipulate. The problem is that with a constantly changing user base, changes you make may not be noticeable.

To get a clear view of how your manipulations have made an impact, you must be able to look at the same group of users over time. In addition to this, engagement rates vary not just between different users but also between the various stages of your users’ lifespans. So optimization will work on some groups but not others, and cohorts allow you to learn where your optimizations have worked.

The definition of cohort analysis is a method of segmenting users from a specific time period to measure key performance indicators (KPIs). Cohort analysis can thereby help reduce churn and/or increase average revenue per user (ARPU).

Cohort analysis can also remove the confusion that can arise from running simultaneous improvements and campaigns, so you can get a clear view of what is yielding the best results and which channels are worth your investment.

The three cohort types for cohort analysis

Today’s app marketers rely on the following three most common types of cohorts to analyze their user or customer cohorts.

1. User acquisition cohorts: Segmented by date of install and channel source.
2. Demographic cohorts: Grouped by characteristics of age, region, gender, etc.
3. Behavioral cohorts: Divided by in-app actions and behavior.

Cohort vs. segment explained

User segmentation is the act of grouping your app users based on shared characteristics like demographics or behaviors. However, in cohort marketing, a time period must be included.

For example, an e-commerce app could build an audience for a marketing message as follows:

Segment: Females who live in New York.

Cohort: New York-based app users who have purchased from the app in the past 30 days.

Upgrading a segment to a cohort in marketing can help further personalize messaging.

Within our dashboard, Adjust clients can gather and define users into customizable cohorts based on install date and/or reattribution date. From this, it’s possible to compare users over time, gauging performance to find out what’s changed and which campaigns result in which users. These dated groups can then be further divided based on the events you’re tracking, such as levels completed or purchases made.

As an added feature, we automatically calculate the lifetime value (LTV) of your users that belong to a cohort. This shows the point at which you break even on paid campaigns.

For more information on cohorts and cohort analysis, take a look at our cohort metric documentation. This resource outlines what cohorts are and how they can be effective, as well as diving into segmentation and your KPI selection. You can also find a full list of terms related to cohort KPIs in the Adjust KPI Service glossary.

To learn how Adjust’s mobile measurement and analytics suite can significantly enhance your cohort marketing, request your personalized demo now.