Cohort Retention
Cohort retention measures how much of a defined customer, user, or recurring-revenue cohort remains at comparable ages after a shared starting event.
The cohort definition, retention basis, starting event, age intervals, and treatment of expansion must all be stated before two curves can be compared.
Cohort retention holds the starting group constant and measures the amount retained at Month 1, Month 3, Month 6, or another comparable age. It can use customers, users, or revenue, but those bases are not interchangeable.
Compare the same starting group at the same ages
This example uses customer-count retention for one fictional cohort of 100 starting customers.
| Cohort age | Customers retained | Interpretation |
|---|---|---|
| Month 0 | 100% | 100 of 100 starting customers |
| Month 1 | 92% | 92 of the original customers remain |
| Month 3 | 84% | 84 of the original customers remain |
| Month 6 | 78% | 78 of the original customers remain |
Compare Month 3 with Month 3 across cohorts, not one cohort’s Month 1 with another cohort’s Month 6.
What is Cohort Retention?
Cohort retention groups customers, users, or revenue by a shared starting event and measures how much remains at comparable ages. A cohort might be defined by signup month, subscription start, first purchase, activation, or another clearly documented event.
The retained amount must match the chosen basis. Customer retention counts retained customers from the original cohort. User retention counts users who meet a defined activity rule. Revenue retention measures recurring revenue associated with the original customer cohort. These curves can move differently.
Age-based columns make cohorts comparable even when they began in different calendar months. Month 3 for a January cohort and Month 3 for a June cohort represent the same relative age, while calendar-quarter totals do not.
Define the cohort and retention event
Starting event
Signup, activation, first purchase, contract start, or another documented event that places an entity in the cohort.
Retention basis
Customer count, active users, units, contracts, or recurring revenue. Name the amount rather than using retention as an unlabeled percentage.
Age interval
A consistent weekly, monthly, quarterly, or annual interval measured from the cohort start.
Retained state
A clear rule for active, paying, renewed, or otherwise retained at each measurement point.
Customer retention versus revenue retention
Customer retention asks how many members of the original cohort remain. Revenue retention asks how much recurring revenue associated with that original cohort remains. Contraction can reduce revenue without losing a customer, while expansion can cause revenue retention to exceed customer retention or even exceed 100% under an NRR-style basis.
Why aggregate retention can hide change
An aggregate rate combines cohorts of different sizes and ages. Strong older cohorts can mask weaker recent acquisition, while a large new cohort can dominate the total before it has aged. A cohort table keeps acquisition period and age visible so the comparison is more diagnostic.
Reading the curve carefully
A falling curve is expected when the definition requires continued activity or payment. The important questions are where the decline occurs, whether recent cohorts differ, how many observations support the rate, and whether product, segment, price, contract, or acquisition mix changed.
Why it matters
Cohort retention separates changes in customer quality or behavior from changes in acquisition volume. It can show whether a newer cohort retains better at Month 3 even when the business-wide customer count is growing or shrinking.
Retention can inform a revenue plan, but a cohort curve is not a cash forecast. Recurring value, invoices, expected collections, and received cash retain separate timing and state. Any plan that uses cohort assumptions should disclose how they enter the model.
Formula with an explicit amount basis
- A fixed cohort membership rule and shared starting event
- A stated amount basis such as customers, active users, or recurring revenue
- Comparable cohort ages and a consistent retained-state rule
- A disclosed treatment of expansion, contraction, pauses, reactivation, and missing observations
Formula with an explicit amount basis
Cohort retention at age t = Retained amount from the original cohort at age t ÷ Starting amount for that cohort × 100
Illustrative Month 0 to Month 6 cohort
A fictional January cohort begins with 100 customers. At Month 1, 92 remain; at Month 3, 84 remain; at Month 6, 78 remain. Customer retention is therefore 92%, 84%, and 78% at those ages.
If the remaining customers expand their subscriptions, revenue retention could be higher than customer retention. That does not restore lost customers. Report the two bases separately and compare this cohort’s Month 3 result with Month 3 for other cohorts.
How RunwayCal helps
RunwayCal can hold explicit revenue and planning assumptions while keeping commercial states and realized cash distinct. Revenue Intelligence supports review of revenue and collection context, and Planner or Scenarios can hold a stated retention assumption.
RunwayCal should not be described as a native cohort-retention analytics product unless a verified product surface supports that claim. The illustrative matrix here explains the concept rather than representing product UI.
Common mistakes
- 1Comparing cohorts at different ages or using inconsistent starting events.
- 2Calling customer, user, and revenue retention the same measure.
- 3Letting expansion obscure customer loss without showing the revenue basis.
- 4Presenting a small cohort percentage without the underlying count or sample context.
Get the Financial Clarity Newsletter
Practical tips on cash flow, runway, and financial decisions for founders, business owners, CFOs, investors, and board members. Free, weekly, no spam.
Turn a retention assumption into an explicit planning input
Keep the cohort basis visible, compare like-for-like ages, and separate recurring-revenue assumptions from realized cash.
Explore Revenue Planning