Churn is quoted as one number and is at least three. Which one you use changes the conclusion entirely.
The three versions
Customer churn = customers lost ÷ customers at the start of the period.
Revenue churn = MRR lost ÷ MRR at the start of the period.
Net revenue churn = (MRR lost − expansion MRR) ÷ MRR at the start.
They diverge whenever customers are different sizes. Lose ten $10 customers and keep one $500 customer, from a base of eleven customers and $600:
- Customer churn: 91%
- Revenue churn: 17%
Both are true. Reporting only one is how a business talks itself into or out of a problem.
Which to watch
Small self-serve, similar plan sizes: customer churn. It is the honest one when revenue per customer is uniform.
Mixed plan sizes or any enterprise tier: revenue churn. Losing your largest account matters more than losing five of your smallest.
Once expansion revenue exists: net revenue churn. Negative net churn - expansion exceeding losses - means revenue grows with no new customers at all, which is the strongest position in SaaS.
What is normal
Rough monthly figures for self-serve products:
| Segment | Monthly customer churn |
|---|---|
| Consumer / prosumer | 5–10% |
| SMB self-serve | 3–7% |
| Mid-market | 1–2% |
| Enterprise | under 1% |
Above 8% monthly for an SMB product means roughly two-thirds of your customers are gone within a year. No acquisition rate outruns that, and every dollar spent on marketing leaks straight back out.
Voluntary versus involuntary
Involuntary churn - failed payments, expired cards - is typically 20 to 40% of total churn and is the cheapest thing on this list to fix. Dunning emails, card-expiry warnings and smart retry schedules recover a large share of it.
If you have never separated the two, do that before working on anything else. It is common to find a third of your churn is a billing problem rather than a product problem.
Churn is not constant
The simple LTV formula assumes a constant churn rate. It is not - churn is highest in the first weeks and falls with tenure.
That is why cohort retention curves are worth more than a single rate. A cohort that loses 40% in month one and 3% a month afterwards has a completely different economic profile from one that loses 8% every month, even though the blended rate can look similar.
If most of your churn is in the first two weeks, you have an onboarding problem, not a retention problem, and they need different fixes.
What actually reduces it
In rough order of effect:
- Fix involuntary churn. Retries and dunning. Pure recovery, no product work.
- Fix the first week. Most cancellations are decided before the customer ever reached the value.
- Sell to the right people. A large share of churn is customers who were never a fit and were sold anyway.
- Add expansion paths. Not churn reduction exactly, but it offsets churn in the net number and is often easier.
- Annual plans. They convert twelve churn decisions into one.
Before optimising anything, split churn into voluntary and involuntary and plot retention by cohort. A single blended monthly percentage hides which of four different problems you actually have.
Launch it where the numbers are checked
RankCert ranks products on domain control we verify ourselves. Listing is free and the link stays dofollow whether or not you display the badge.