Cohort Analysis
Cohort analysis groups customers by when they started (e.g. all January signups) and tracks each group's retention and revenue separately over time. It is the only way to see whether retention is actually improving, because blended averages mix old and new customers.
Formula
Worked example
The January cohort retains 71% of revenue at month 6; the July cohort, after onboarding changes, retains 84% at month 6. Blended churn barely moved — the cohort view shows the fix worked.
The standard artefact is a cohort triangle: rows are start months, columns are months since start, cells are retention percentages. Read down a column to compare cohorts at the same age; read along a row to see how one cohort decays.
Cohorts also expose where the naive LTV formula lies: if retention curves flatten after month 12 (they usually do), constant-churn LTV understates mature-cohort value and overstates early decay.
Compute it: Churn rate calculator