Cohort Analysis
A method of grouping users by a shared attribute or date and tracking their behavior over time.
Definition
Cohort analysis is a technique for understanding user behavior by grouping users who share a characteristic - usually their first-install date or signup date - and tracking what they do over time as a group. By comparing cohorts against each other, you can isolate the effect of product changes: if the March cohort retains better than the February cohort, a change shipped between the two dates likely caused it. For mobile apps, cohort analysis is the standard method for measuring retention because it prevents newer users from masking the behavior of older ones.
→ In practice
"We ran a cohort analysis on users who signed up in Q4 vs Q1. The Q1 cohort had 30% better day-30 retention - which correlates with the onboarding redesign we shipped in January."
Common questions
What is a behavioral cohort vs a date cohort?
A date cohort groups users by when they signed up or installed. A behavioral cohort groups users by something they did - e.g., "users who connected Stripe in their first week." Behavioral cohorts are useful for comparing activated vs non-activated users.
How long should a cohort analysis look-back window be?
Choose a window that matches the product's natural return cycle and the decision you need to make. Keep it fixed while comparing cohorts, and report the cohort size and acquisition period alongside the result.
Related terms
Track cohort analysis for your app
See your onboarding metrics live
OnRamp tracks cohort analysis and other key onboarding metrics automatically - step funnel, retention cohorts, and session timelines - without a data team.
Start free trial →