What Is Cohort Analysis?
Cohort analysis groups customers by when they joined and tracks how they behave over time. It's one of the most powerful tools for understanding retention.
Key Takeaways
- A cohort is a group of customers who share a common characteristic — typically the month they first purchased.
- Cohort analysis tracks how each group's behaviour evolves over time.
- It reveals whether retention is improving, worsening, or staying the same across different customer generations.
What a cohort is
In customer analytics, a cohort is a group of customers who share a common starting point — typically the month or quarter they made their first purchase. Cohort analysis tracks how that group's behaviour (purchase rate, spending, retention) changes over subsequent months. Each month's new customers form a new cohort.
How to read a cohort table
A cohort retention table shows cohorts as rows (January customers, February customers, etc.) and months since first purchase as columns. The first column always shows 100% (month 0 = acquisition). Subsequent columns show what percentage of that cohort returned to buy again. Comparing rows reveals whether newer customers retain better or worse than older ones.
What cohort analysis reveals
Cohort analysis answers questions aggregate retention rates cannot. Are customers acquired through a specific channel retaining better? Did a product change improve or damage retention for cohorts acquired after the change? Is there a seasonal pattern in the quality of acquired customers? These insights are invisible in aggregate but clear in cohort data.
Practical application
If your January cohort shows 40% retention at month 3, and your June cohort (acquired after a product improvement) shows 60% at month 3, the improvement has measurably increased retention. You can now attribute business value to product decisions — and identify which decisions had the biggest positive impact on long-term customer quality.