What Is Churn? A Plain English Guide for SaaS
Churn is the rate at which customers or revenue leave your business. Here's what it means, why it matters, and what you can actually do about it.
Churn, simply put
Churn is when customers leave. In a subscription business, that means a paying customer stops being a paying customer, whether they cancelled on purpose, their payment failed, or their contract expired and they didn’t renew.
The word gets used loosely. “We’re churning too many customers” and “our churn rate is 6%” mean different things. The first is a general observation. The second is a specific metric. Both matter, but the metric is what you can actually act on.
Types of churn
Not all churn is the same, and the distinction changes what you should do about it.
Customer churn vs revenue churn
Customer churn counts the number of customers who left. Revenue churn counts the money that left. They often tell different stories.
Say you lose five customers in a month. If they were all on your $19 plan, that’s $95 in lost MRR. If one was on your $499 enterprise plan and four were on the $19 plan, that’s $575. Same customer churn, very different revenue impact.
Revenue churn can also go negative. If your existing customers upgrade or expand by more than the revenue you lost to cancellations, you’ve got negative churn, which is effectively a growth engine powered by your installed base.
Voluntary vs involuntary
This is the distinction that changes your strategy completely.
Voluntary churn is a customer actively choosing to leave. They cancelled because they found a better alternative, outgrew your product, or never saw enough value. Fixing voluntary churn requires product work, better onboarding, customer success, or pricing changes. It’s important, and it’s slow.
Involuntary churn is a customer lost to a failed payment. Their card expired, their bank declined the charge, or they hit a temporary balance issue. The customer didn’t choose to leave. Recurly’s data suggests involuntary churn accounts for 20-40% of total churn across subscription businesses.
Here’s why the distinction matters: involuntary churn is a mechanical problem with a mechanical solution. Automated dunning, smart retries, and pre-dunning alerts for expiring cards can recover the majority of these failures. No product changes needed, no lengthy customer success initiatives. It’s the fastest retention win available.
Why churn matters for SaaS
Churn compounds. That’s the part most founders underestimate.
At 5% monthly churn, you’re not just losing 5% of customers each month. Over a year, you’d lose about 46% of your starting customer base (the compound formula is 1 − (1 − 0.05)^12). That means nearly half your customers need to be replaced every year just to stay flat.
The financial impact cascades. Higher churn means lower customer lifetime value, which means you can afford to spend less on acquisition, which means slower growth, which makes fundraising harder. It’s a cycle that’s difficult to break once it sets in.
On the other side, small improvements compound too. Reducing monthly churn from 5% to 4% improves your annual retention from 54% to 61%. That single percentage point change means keeping an additional 7% of your customer base every year. Plug your numbers into our churn rate calculator to see the impact for your business.
How churn rate is calculated
The basic formula:
Churn rate = (Customers lost during period / Customers at start of period) × 100
Start the month with 400 customers, lose 20, and your monthly churn rate is 5%.
For revenue churn, replace customer counts with MRR:
Revenue churn rate = (MRR lost to cancellations and downgrades / Starting MRR) × 100
Both formulas look simple, but the details trip people up. Should you count customers who signed up and cancelled within the same period? What about paused subscriptions? Most SaaS analytics tools (Stripe, Baremetrics, ChartMogul) handle these edge cases differently, which is why your churn rate can vary depending on which dashboard you’re looking at.
What is a good churn rate?
Benchmarks vary by segment, contract type, and how you’re measuring.
| Segment | Typical monthly churn | Notes |
|---|---|---|
| Enterprise B2B (annual contracts) | 0.5-1% | High switching costs, dedicated CSMs |
| Mid-market B2B | 2-3% | Mix of annual and monthly billing |
| SMB B2B (monthly billing) | 5-8% | Lower switching costs, price-sensitive |
| B2C subscriptions | 6-10% | High volume, lower engagement |
These are rough guides. In practice, your churn rate depends on your pricing, your product’s stickiness, your market’s maturity, and whether you’re measuring customer churn or revenue churn. A 5% customer churn rate where you’re only losing low-tier customers is a completely different situation from 5% where enterprise accounts are leaving.
We’ve found the most useful benchmark is your own trend line. If churn is declining quarter over quarter, you’re heading in the right direction regardless of where you sit against industry averages.
How to reduce churn
The approach depends on which type of churn you’re dealing with.
Fix involuntary churn first. It’s the fastest win and requires the least organisational effort. Set up automated payment retries, send pre-dunning alerts before cards expire, and build a recovery email sequence for failed payments. Stripe’s Smart Retries recover about 57% of failed payments on their own, but dedicated recovery tools push that higher.
For voluntary churn, the levers are broader and slower. Improve onboarding so customers reach their “aha moment” faster. Monitor engagement signals and intervene when usage drops. Review your pricing to ensure it aligns with the value customers are getting. Run exit surveys to understand why people actually leave, not just why they say they leave.
One pattern we’ve seen work: segment your churn by cohort, plan tier, and acquisition channel. The answer to “why are customers leaving?” is almost never a single reason. SMB customers on your cheapest plan might be churning because of price. Enterprise customers might be leaving because of a missing integration. Aggregated churn hides these signals.