What is churn rate?
Churn rate is the share of customers, users or revenue lost during a period. It is calculated by dividing the number lost by the number at the start of the period, excluding anyone acquired during it. Logo churn counts customers; revenue churn counts money, and the two can move in opposite directions.
Formula
Churn rate = (lost during period / total at start of period) x 100
- lost during period
- Customers who cancelled or lapsed. Do not include anyone who joined mid-period.
- total at start
- The count on day one of the period. Using an average balance instead inflates the denominator and understates churn.
Worked example
A subscription product in September.
| Customers on 1 September | 820 |
|---|---|
| Cancelled during September | 31 |
| New customers in September | 94, excluded from the calculation |
| MRR on 1 September | $41,000 |
| MRR lost to cancellations | $980 |
| Expansion MRR from existing customers | $1,640 |
Logo churn = 3.8%, gross revenue churn = 2.4%, net revenue churn = -1.6% Nearly 4% of customers left, yet revenue from the existing base grew. The small accounts are leaving and the large ones are expanding, which is a different business problem from the one logo churn alone describes.
Logo churn, gross revenue churn, net revenue churn
Three metrics, one word, and quoting the wrong one is how board decks get misread.
- Logo churn counts customers, treating a $9 account and a $9,000 account identically. It is the right lens on product-market fit.
- Gross revenue churn counts money lost from cancellations and downgrades. It is what the finance team means by churn.
- Net revenue churn subtracts expansion from existing customers. Below zero, sometimes called negative churn, means the existing base grows on its own even with no new sales.
A healthy SaaS business can run 4% logo churn and negative net revenue churn at the same time. Reporting only the first number makes it look like it is bleeding; reporting only the second hides that the low-end product is not working.
Compared
Monthly churn compounds fast The reason small monthly numbers matter: the survival rate is multiplicative, so an apparently modest monthly rate decides how much of a cohort exists a year later.
| Monthly churn | Retained after 12 months | Average customer lifetime |
|---|---|---|
| 1% | 89% | About 100 months |
| 3% | 69% | About 33 months |
| 5% | 54% | 20 months |
| 8% | 37% | 12.5 months |
| 12% | 22% | About 8 months |
Retained after 12 months is (1 - churn) to the power of 12. Average lifetime is 1 / churn, which assumes a constant rate and so overstates lifetime for products whose churn falls as accounts mature.
Live demo
See it on real data. The live demo is a fully populated dashboard on demo traffic, with no signup and no card.
Open the live demoWhere churn numbers go wrong
Counting new customers in the denominator
Adding mid-period signups inflates the base and quietly understates churn. Measure against the count at the start of the period.
Annual plans hidden in a monthly rate
Annual subscribers can only churn on their renewal date, so a mixed book produces churn spikes twelve months after every good sales quarter. Segment by billing term.
Involuntary churn ignored
Expired cards and failed payments are often a large share of cancellations, and they respond to dunning emails rather than to product work. Split them out before drawing product conclusions.
No leading indicator
Churn is confirmed weeks after the customer stopped caring. Declining usage frequency, in your product analytics, is the signal that actually arrives in time to act on.
Seeing churn before it is billed
By the time a cancellation is recorded, the decision is old. The useful work happens upstream, in behaviour: accounts whose weekly active usage halved, accounts that stopped using the one feature that correlates with renewal, accounts where the champion stopped logging in.
That is a cohort question, and it is answerable with the same data as retention. Build a cohort of accounts by signup month, watch the activity curve, and the accounts sliding toward the floor are your churn pipeline.
mrkr does not bill your customers, so it does not know who cancelled. Its event and funnel reports show recorded activity; they do not calculate customer churn or save cohort groups.
FAQ
What is a good churn rate for SaaS?
For SMB-focused products, 3% to 5% monthly logo churn is common. For enterprise contracts, under 1% monthly is the usual expectation. Net revenue churn below zero, meaning expansion outweighs losses, is the strongest signal a subscription business can show.
How do I calculate annual churn from monthly churn?
Compound it rather than multiplying by twelve. Annual churn is 1 minus (1 minus monthly churn) to the power of 12. At 5% monthly that is 46% annually, not 60%.
What is negative churn?
Net revenue churn below zero: upgrades and expansion from existing customers exceed the revenue lost to cancellations and downgrades, so the existing base grows without any new customers. It requires a pricing model where accounts naturally pay more as they get more value.
Start with one line of HTML.Your first visitors show up within a minute.