Metrics
Cohort Analysis
Cohort analysis groups customers by a shared starting point — typically the month they signed up — and tracks how that specific group's behaviour (retention, spend, engagement) changes over subsequent months, rather than looking at the whole customer base's aggregate metrics in each period.
Aggregate metrics can mask real problems in a growing company: if new customer volume is rising fast enough, a growing total customer count and growing total revenue can coexist with each individual cohort retaining worse than the one before it. Cohort analysis is often the only way this shows up before it becomes a serious problem.
The standard output is a retention curve or triangle table: each row is a monthly cohort, each column is months since signup, and the values show what percentage of that cohort is still active (or how much revenue it still generates) at each point — letting later cohorts be compared against earlier ones at the same age.