Churn Rate Calculator with Cohort Retention Heatmap
Work out customer churn, revenue churn, NRR, GRR and average customer lifetime — then map every monthly cohort on a real retention heatmap instead of settling for one bare percentage. Runs entirely in your browser.
Customer churn — counting logos
How many accounts left, regardless of what they paid you.
Everyone paying you on day one of the period.
Accounts that cancelled during the period.
Signups in the period. Never counted in retention.
Customer churn rate
4.0%
48 of 1,200 logos
Customer retention rate
96.0%
100% minus churn
Avg customer lifetime
25.0 mo
1 ÷ monthly churn
Customers at end
1,282
start − lost + new
Revenue churn — counting dollars
How much recurring revenue left, regardless of how many accounts it came from.
Recurring revenue on day one, from those same accounts.
Cancellations plus any plan downgrades.
Upgrades, seat adds and overages from existing accounts.
Gross revenue churn
5.0%
MRR lost ÷ starting MRR
Net revenue churn
-1.3%
negative — expansion wins
Gross revenue retention
95.0%
excludes expansion, caps at 100%
Net revenue retention
101.3%
growing without new logos
These two are not the same number. Yours are 4.0% customer churn against 5.0% revenue churn — a 1.0pt gap, meaning the accounts you lost were larger than your average. Quoting only the customer number would understate the damage.
Cohort retention heatmap
Set each cohort size, then type the customers still active in each later month. The small number inside a cell is the headcount you edit; the big number is the retention it implies.
| Cohort | Started | M0 | M1 | M2 | M3 | M4 | M5 |
|---|---|---|---|---|---|---|---|
| Jan | 100%320 | 85% | 75% | 69% | 66% | 63% | |
| Feb | 100%410 | 86% | 77% | 71% | 68% | not yet | |
| Mar | 100%380 | 87% | 79% | 74% | not yet | not yet | |
| Apr | 100%465 | 88% | 81% | not yet | not yet | not yet | |
| May | 100%520 | 90% | not yet | not yet | not yet | not yet | |
| Jun | 100%610 | not yet | not yet | not yet | not yet | not yet | |
| Average | 2,705 | 100%6 cohorts | 87%5 cohorts | 78%4 cohorts | 72%3 cohorts | 67%2 cohorts | 63%1 cohort |
The churn formulas, written out
Every figure this churn rate calculator produces comes from one of the seven formulas below. All of them run over the same period — normally one month — and none of them count customers you acquired during that period, because acquisition is not retention.
Customer churn rate
churned customers ÷ customers at start × 100New customers won during the period are excluded from both sides. Including them flatters the number.
Customer retention rate
(customers at end − new customers) ÷ customers at start × 100The exact mirror of customer churn. The two always add up to 100%.
Gross revenue churn
MRR lost to cancellations and downgrades ÷ MRR at start × 100Expansion revenue is deliberately ignored, so this can never be negative.
Net revenue churn
(MRR lost − expansion MRR) ÷ MRR at start × 100Goes negative when your existing base grows faster than it leaks. Negative net churn is the goal.
Gross revenue retention (GRR)
(MRR at start − MRR lost) ÷ MRR at start × 100Capped at 100% by construction. It measures how well you hold what you already sold.
Net revenue retention (NRR)
(MRR at start − MRR lost + expansion MRR) ÷ MRR at start × 100Uncapped. Above 100% means the existing base alone grows revenue with zero new logos.
Average customer lifetime
1 ÷ monthly customer churn rateA 4% monthly churn implies roughly 25 months. It assumes a constant churn rate, which cohorts rarely obey.
Why revenue churn and customer churn diverge
Customer churn treats every account as worth exactly one. Revenue churn weights each account by what it pays. The two only agree in the imaginary business where every customer is on the same plan, and conflating them is the single most common mistake in churn reporting.
Suppose you start a month with 100 customers and $50,000 of MRR, and five leave. If those five were all on your $99 starter plan, customer churn is 5% but revenue churn is under 1% — the business is barely scratched. If instead one of the five was a $20,000 enterprise account, customer churn is still 5% while revenue churn is over 40%. Same logo count, completely different quarter.
The gap between the two numbers is itself the signal. Revenue churn above customer churn means you are losing your bigger accounts, which usually points at onboarding, support coverage or a missing account manager. Revenue churn below customer churn means your leavers skew small, which is typically a pricing or qualification problem at the bottom of the funnel — annoying, but far less dangerous. Always state which of the two you are quoting.
Why NRR above 100% is possible
Gross revenue retention only ever subtracts. It starts from your opening MRR, removes cancellations and downgrades, and stops there — so it is mathematically capped at 100%. It answers one question: how much of what you already sold did you keep?
Net revenue retention adds expansion back in — upgrades, extra seats, usage overages, cross-sells. Because that term is additive and unbounded, NRR can and often does exceed 100%. An NRR of 115% means that if you had signed zero new customers, revenue from your existing base would still have grown 15% over the period. That is why investors read NRR as a proxy for whether the product gets more valuable the longer someone uses it.
Report them as a pair. A company with 120% NRR and 82% GRR is expanding a leaky base — a few large accounts are masking real churn underneath. A company with 105% NRR and 96% GRR is far healthier, even though the headline number is lower. NRR without GRR beside it hides exactly the problem you most need to see.
How to read a cohort retention triangle
A single churn percentage is an average across customers who joined at wildly different times under wildly different versions of your product. A cohort table takes that average apart. Each row is one month of signups, followed forward in time; each column is an age, not a date. The grid is triangular because your newest cohort has only lived one month, so there is nothing to put in the later columns yet.
Read it two ways. Across a row is the survival curve for one cohort: expect a steep drop from M0 to M1 or M2, then a flattening tail. If the curve never flattens, you have a product problem, not an onboarding one — customers keep deciding to leave rather than failing to get started. Down a column compares cohorts at the same age, which is the only honest way to tell whether last quarter of work actually improved retention. If M3 reads 68%, 71%, then 75% as you scan down, newer cohorts are retaining better and something you changed is working.
The average row at the bottom of the grid collapses each column into one figure, weighting every cohort equally. Treat the later columns with care: they are averaged over fewer cohorts, so one unusual month moves them a lot. The cell caption tells you how many cohorts each average is built from.
Churn and retention — frequently asked
What is a good monthly churn rate?
For self-serve SMB SaaS, 3–5% monthly customer churn is normal and under 3% is strong. Mid-market lands nearer 1–2%, and enterprise contracts are usually measured annually rather than monthly, where under 10% a year is the benchmark. Compare yourself to your own segment: a 5% monthly churn is unremarkable on a $29 plan and alarming on a $2,000 one.
How do I calculate customer retention rate?
Take the customers you finished the period with, subtract any you acquired during it, divide by the customers you started with, then multiply by 100. Subtracting new signups is the step most people skip, and skipping it lets acquisition disguise a retention problem. Retention rate and churn rate are the same measurement from opposite ends, so they sum to 100%.
What is the difference between revenue churn and customer churn?
Customer churn counts logos leaving; revenue churn counts dollars leaving. They only match when every customer pays the same amount. If your churn is concentrated among small accounts, revenue churn comes in below customer churn. If a handful of large accounts leave, revenue churn spikes while customer churn barely moves. Track both, and always say which one you mean when quoting a churn figure.
Can net revenue retention be over 100%?
Yes, and that is the point of measuring it. NRR includes expansion revenue from upgrades, extra seats and usage overages, so when the surviving customers spend more than the leavers took away, NRR rises above 100%. Best-in-class B2B SaaS reports 110–130%. Gross revenue retention excludes expansion and therefore can never exceed 100%, which is why the two are always reported as a pair.
How many months of data do I need for a cohort retention chart?
Six monthly cohorts is enough to see a shape, which is why the grid above starts with six. The oldest cohort gives you a long curve from a single group, and the newest ones show whether recent changes helped. Read the average row at the bottom for the overall curve, and read down each column to see whether newer cohorts retain better than older ones at the same age.
Keep going
Browse all free tools.