Cohort analysis

Cohort analysis groups customers by something they share — usually the month of their first purchase — and tracks each group over time rather than blending everyone into one average. Lifecycle teams use it to separate a genuine retention problem from a change in acquisition mix, which a blended repeat rate hides. The common mistake is reading young cohorts too early: a 30-day window flatters cohorts whose typical repurchase gap is 90 days.