Cohort retention calculator. With 24-month survivor curve.
Enter starting cohort size and monthly retention rate. Get the survivor curve at month 1, 3, 6, 12, 24, plus the cohort half-life and the chart of monthly decay.
Starting size, monthly rate.
Survivors + half-life.
Cohorts tell the truth.
Cohort retention measures the percentage of a single signup-period's users who remain active in subsequent months. The cohort approach corrects for the survivorship bias of aggregate retention numbers, which can show flat retention even while new cohorts decay faster. The calculator above projects a cohort over 24 months using constant monthly retention, computes half-life (the months until cohort is halved), and shows survivors at month 1, 3, 6, 12, 24 against category benchmark. Half-life is the cleanest single-number summary of retention quality across categories.
Three principles for retention work. One, fix month 1-3 first — Andrew Chen and Reforge research consistently show that 60-80% of total retention loss happens in the first 90 days from incomplete activation. Two, watch the curve shape, not the M12 number alone — a curve that flattens by month 6 (still showing 50%+ retention) signals product-market fit; one that decays linearly past month 12 signals weak fit even at high M1. Three, segment cohorts by acquisition channel — the same M12 retention can hide great paid cohorts and weak organic cohorts (or vice versa); aggregating the two kills the diagnostic signal.
Tools in the same cluster: Churn Rate Calculator for the inverse view (loss instead of survival). LTV Calculator for the lifetime-revenue projection from retained cohort. SaaS Quick Ratio for the new-vs-churn financial-efficiency view.
Six answers.
What is cohort retention?
Cohort retention measures the percentage of users from a single signup period (typically a month) who remain active across subsequent periods. A cohort of 1,000 users signed up in January, with 850 still active in February (M1 retention 85%), 720 in March (M2 72%), 540 in June (M5 54%) tells the story of natural decay. The cohort approach corrects for the survivorship bias of aggregate retention numbers and exposes whether retention is improving, flat, or worsening across new-user generations.
What's a healthy retention curve by category?
B2B SaaS: M1 retention 85-92%, M12 retention 70-80% (best-in-class hits 90%+). Consumer SaaS: M1 60-70%, M12 35-45%. Mobile app (consumer): M1 35-45%, M12 5-15% (the steep mobile decay curve). DTC subscription (beauty, supplements): M1 65-80%, M12 25-40%. Marketplace (two-sided): demand-side M1 50-65%, M12 30-45%; supply-side M1 70-85%, M12 50-65%. Curves that flatten by month 6 (still showing 50%+ retention) signal product-market fit; curves that decay linearly past month 12 signal weak fit even at high M1.
What is cohort half-life?
Half-life is the number of months until the cohort is reduced by half. A 90% monthly retention cohort has a half-life of about 6.6 months; an 80% retention cohort has a half-life of 3.1 months; a 95% retention cohort has a half-life of 13.5 months. Half-life is the cleanest single-number summary of retention quality because it is interpretable across categories and stages. The calculator above computes half-life automatically — useful for back-of-envelope LTV math.
Why do retention curves usually flatten over time?
Self-selection. Users who churn early tend to be wrong-fit (acquired via the wrong channel, didn't activate, didn't see value). Users who survive past month 6 tend to be product-fit users who built the product into their workflow or routine. Their forward retention rate is meaningfully higher than the early cohort's. Most healthy SaaS curves bend toward an asymptote (the long-term retained-fraction) somewhere between months 6-12. If the curve never flattens, the product has weak fit even with the apparent loyal cohort.
Smooth or stepwise retention modeling?
The calculator above uses constant monthly retention rate, which produces a smooth exponential decay. Real-world cohorts have step-function retention (large drops at month 1 from non-activation, then smaller drops). For board reporting, run actual cohort tables monthly and average across 6 cohorts to smooth noise. For modeling, the constant-retention exponential is the standard simplification — it overstates short-term retention and understates long-term retention but matches the steady-state by month 12 for most products.
Does this tool save my data?
No. Every value lives in this browser tab only. Nothing is sent to any server. Closing the tab clears the data. The Copy Results button puts a plain-text summary on your clipboard.
M12 retention below category?
Our SaaS development engagements ship the activation flow, in-product onboarding, and lifecycle-email stack that lifts month-1 to month-3 retention — the highest-ROI retention work in any SaaS.
Published .