Cohort Retention Calculator

A cohort is a group of customers or users who share a common starting event, such as the month they first purchased, the week they activated their account, or the date they entered a program. Cohort-based retention tracks what fraction of that original group remains active as time passes, instead of averaging behavior across all customers.

In the period cohort retention method, the calculator divides the number of active customers in period N by the original cohort size. If you started with 1,000 customers in the January cohort and 420 are still active in month 6, then month-6 retention for that cohort is 420 ÷ 1,000 = 0.42, or 42 percent, and the implied churn for that period is 1 minus 0.42, or 58 percent.

In the rolling retention method, you count customers who are active at or after a given period N, then divide by the starting cohort size. This view is often used in product analytics because it treats any customer who comes back at or beyond the target period as retained, even if they skipped an intermediate period.

The cohort survival view in this tool starts from an estimate of the number of customers who have churned by the end of period N and backs out the survival rate as one minus churn. This is useful when your source data directly tracks losses or cancellations rather than active users, such as in credit-risk, medical follow-up, or subscription cancellation datasets.

All three methods are mathematically consistent with standard treatments of cohort retention in analytics and social-science research: they express retention as the number of individuals remaining at a given follow-up time divided by the number at risk at the start, and express churn as the complement of retention.

To use the calculator responsibly, define cohorts on a consistent basis, use the same time bucket across cohorts, and avoid over-interpreting very small cohorts, where a change of a few users can produce large swings in the retention percentage. For operational use, pair this calculator with a retention table in your warehouse or BI tool so you can validate numbers against the underlying data.