How do you calculate cohort retention rate in a spreadsheet?

Export your user list with signup date and last active date. Group users by signup month (cohort). For each cohort, count how many users were active in month 1, month 2, etc. after signup. Divide each month's active count by the original cohort size and multiply by 100 to get the retention percentage. Month 0 is always 100%.

What is a good month-3 retention rate for a B2B SaaS product?

A good month-3 retention rate for B2B SaaS is 40–55% for SMB-focused products and 65–75% for enterprise-focused products. Top-quartile B2B SaaS products retain 70%+ of each cohort through month 3. If you are below 30% at month 3 for any cohort, that is a strong signal of an activation or value delivery problem.

What is the difference between cohort retention and rolling retention?

Cohort retention measures whether a user was active specifically within a given period after signup (bounded). Rolling retention counts a user as retained if they are active at any point on or after that period (unbounded). Rolling retention always produces higher numbers and is less useful for identifying habitual usage patterns. Subscription businesses should use bounded cohort retention aligned to billing cycles.

How many users do you need for cohort retention analysis to be meaningful?

You need at least 30–50 users per cohort for the percentages to be statistically meaningful. With fewer than 30 users in a cohort, a single user's behavior moves the retention rate by 3+ percentage points, making trends unreliable. For early-stage products with small cohorts, use quarterly cohorts instead of monthly to aggregate enough users.

Why do later cohorts sometimes retain better than earlier ones?

Later cohorts often retain better because the product has improved through iteration, onboarding has been refined based on early user feedback, and marketing has become more targeted toward high-fit customers. This is a healthy pattern and one of the key signals to look for in your cohort matrix—it confirms that product improvements and acquisition targeting changes are having a measurable retention impact.

How to Do Cohort Retention Analysis

What Cohort Retention Analysis Measures

A cohort is a group of users who share a defining event within the same time window—most commonly their signup date. Cohort retention analysis asks: of the users who started in period X, what percentage was still active in period X+1, X+2, and so on?

This is more actionable than aggregate retention because it separates the signal from different acquisition periods. If your December cohort retains at 60% through month 3 and your March cohort retains at only 30%, something changed—a product update, a pricing change, a channel shift—between those two months. An aggregate number would hide that entirely.

Step-by-Step: Running a Cohort Retention Analysis

Step 1: Choose your cohort definition. For SaaS, this is almost always signup month (or week for fast-growth products). For mobile apps, it is often install date. The cohort definition must be consistent—mixing acquisition channels into a single cohort obscures channel-level variance.

Step 2: Define your retention event. 'Active' must mean something specific. Options include: logged in, completed a core action (ran an analysis, published a report, sent a message), or paid. For SaaS, the gold standard is a product-specific action tied to value delivery, not just a login event. Login-based retention overstates engagement by 20–40% compared to action-based retention in most SaaS products.

Step 3: Build the retention matrix. Rows are cohorts (Jan, Feb, Mar…). Columns are periods after acquisition (month 0, month 1, month 2…). Each cell contains the percentage of the cohort still active in that period. Month 0 is always 100% by definition.

Step 4: Look for the retention curve shape. Most SaaS products show a steep drop in months 1–2 as non-activated users churn, then a flattening curve as retained users stabilize. Where the curve flattens is your retained core—users who have found durable value. Products with no flattening curve (continual decline) have a fundamental value problem.

Retention Benchmarks by SaaS Segment

Segment	Month 1 Retention	Month 3 Retention	Month 6 Retention	Source
B2B SaaS (SMB)	65–75%	40–55%	30–45%	Amplitude Benchmarks
B2B SaaS (Enterprise)	80–90%	65–75%	55–70%	ProfitWell
B2C SaaS / Consumer	25–40%	12–25%	8–18%	Mixpanel Benchmarks
Mobile Apps	20–35%	8–15%	4–10%	AppsFlyer 2024
Top-quartile B2B SaaS	85%+	70%+	60%+	Amplitude Benchmarks

Reading the Retention Matrix: What to Look For

Three patterns reveal specific problems:

High drop at month 1: Activation failure. Users sign up but don't reach the moment of value. This is an onboarding problem, not a product problem.
Consistent decline with no flattening: No retained core. The product delivers value inconsistently, or the use case doesn't recur often enough. Common in tools users only need occasionally.
Cohort degradation over time: If each successive cohort retains worse than the previous one, something systemic is changing—pricing, onboarding, product quality, or customer fit from marketing channels.

Cohort degradation is the most dangerous pattern and the hardest to see without cohort analysis. It often signals that a growth channel is acquiring lower-quality users, but it can also signal product regression.

Cohort Analysis vs. Rolling Retention

Rolling retention (also called unbounded retention) counts a user as retained in a period if they are active at any point on or after that period—not just within it. This overstates retention. Bounded cohort retention—active specifically within period N—gives a more accurate picture of habitual usage. For subscription businesses, use bounded retention aligned to billing cycles.

Tools for Cohort Analysis

Most product analytics platforms support cohort retention out of the box: Amplitude, Mixpanel, PostHog, and Heap all provide configurable retention tables. For subscription metrics specifically, Baremetrics and ChartMogul compute cohort revenue retention. At small scale, a simple spreadsheet with monthly user export data is sufficient for cohorts of under 500 users per period.

For SaaS teams that also track cancellation feedback, cohort analysis pairs directly with exit survey data. A cohort that churns heavily at month 2 can be cross-referenced with cancellation feedback analysis to identify the specific friction causing that drop.

If you are new to the underlying retention math, the churn rate formula guide explains how cohort retention rates convert to monthly and annual churn figures. And if your cohorts show improving retention over time, that is the early signal of the churn prediction inputs worth monitoring to catch deterioration early.