Cohort analysis is a method of grouping customers or users who share a characteristic, most often the period they signed up or purchased, and comparing how each group behaves over time, for example how many are still active after one, two or three months.
Why it matters
Blended averages can hide what is happening. If a product keeps adding new users, overall retention may look stable while each new cohort retains worse than the last. Cohort tables expose such changes and show whether product improvements, pricing changes or marketing campaigns actually affected behaviour. They apply to retention, revenue per customer, repeat purchases and churn.
Example
Share of each signup cohort still active:
Cohort Month 0 Month 1 Month 2 Month 3
Jan 100% 48% 38% 34%
Feb 100% 51% 41% 37%
Mar 100% 57% 47% 43%Later cohorts hold better, which may reflect an improved onboarding flow introduced in February.
Common mistakes
- Comparing cohorts of very different sizes without noting it.
- Defining "active" inconsistently between reports.
- Reading young cohorts as final, when they have fewer months of data.
Related
Get notified when this term updates
Plus occasional practical tips on uptime and monitoring, nothing else. Unsubscribe anytime.
This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.