What is Cohort Retention?

Cohort retention shows how many customers from the same signup month are still paying you later. It helps SaaS teams see whether customers are staying, leaving early, or leaving slowly over time.

That matters because aggregate churn can hide what’s really happening. Your overall churn rate may look flat while one signup group is doing well and another is losing customers fast.

Cohort retention separates those groups, so you can see where retention is getting better or worse.

Instead of asking, “What share of all customers stayed this month?” cohort retention asks, “Of the customers who joined in January, how many are still here in April?”

That’s what makes cohort retention a clearer way to measure whether retention is improving, getting worse, or holding steady.

Key takeaways

• Cohort retention groups customers by signup month and tracks what percentage stays each month after signup.
• The formula is: customers still active at Month N divided by the Month 0 cohort size, times 100.
• Aggregate churn rate can hide trends that cohort retention makes clear.
• 3 common curve shapes are Steep-Then-Flat, Steady-Decline, and Recovery-Curve. Each shape points to a different root cause.

What is the cohort retention formula?

The cohort retention formula divides the number of customers still active at Month N by the number of customers in the cohort at Month 0. Then you multiply that number by 100.

Month 0 is always 100% because every customer who joined is active on day one. Every month after Month 0 shows what percentage of that original group is still active.

Here’s the formula for cohort retention:

(Customers from cohort still active at Month N ÷ Customers in cohort at Month 0) × 100

And here’s an example of this in practice:

In this example, 200 customers joined in January. By April, 140 were still paying. That means cohort retention at Month 3 is 70%.

For the revenue cohort version, replace customer count with monthly recurring revenue (MRR). A cohort can lose customers and still show 110% retention if expansion MRR grows faster than the lost revenue.

To run the numbers on your own data, the Churn.io cohort analysis tool builds the retention matrix automatically.

David Skok’s SaaS Metrics 2.0 shows this with Zendesk’s cohort data. The revenue curves bend upward even while some customers churn, which is one of the clearest examples of negative churn.

What cohort retention reveals that churn rate hides

Aggregate monthly churn rate groups all active customers together, no matter how long they’ve been subscribed. A 3% monthly churn rate tells you that 3% of your customer base canceled.

Aggregate monthly churn doesn’t tell you where the churn is coming from. That 3% may come mostly from customers in their first 30 days, which points to an onboarding problem. Or it may be spread evenly across all customers, which points to a product-value problem.

Those two problems need different fixes, and aggregate churn rate can blur them together.

In our customer base, the majority of churn for new SaaS accounts concentrates in the first 60 days. That pattern is hard to see in aggregate churn rate, but it’s clear in a cohort retention table:

Common misuse: cohort retention vs. aggregate retention rate

Your churn rate can hold flat at 3% while your newest cohorts are churning at 7% and your two-year cohorts are churning at 0.5%. The averages cancel out, so the bad trend stays hidden.

That matters because you may be missing a growing onboarding problem. ChartMogul’s SaaS Retention Report shows how cohort-level data can surface gaps that aggregate metrics hide.

Cohort retention is a measurement method. It’s separate from the 3 retention types: classic retention, range retention, and bracket retention.

Those retention types describe different time windows for counting who returned to your product. Cohort retention describes how you group the users you’re measuring.

Related terms