How to use the subscription cohort retention calculator
A cohort analysis groups only the people who signed up in the same month and follows how many of them are still around later. Total member counts hide churn behind new signups; grouping by signup month shows exactly when people leave.
The table shows two different retention numbers. Cumulative retention divides the people remaining by the original cohort size, while month-over-month retention divides by the previous month. Different denominators mean different numbers, and the familiar retention curve is the cumulative one.
Enter the people still subscribed at each point, not the number who cancelled. Leave future months blank and only the filled months are used. If a month has more people than the month before, win-backs are probably mixed in, so the calculator stops and says so.
Frequently asked questions
It is the geometric mean: the cumulative retention at the last filled month raised to the power of one divided by the number of months. A simple average of monthly rates would understate the steep drop in the first months.
This calculator only handles a cohort that shrinks, so returning subscribers are easier to read when tracked separately rather than added back into the original cohort.