marketing
Churn Rate
Percentage of customers who leave in a given period. The silent killer of recurring revenue.
Definition
Churn rate is customers lost in a period divided by customers at the start of the period. 5% monthly churn means losing 5% of customers every month - which compounds to ~46% annual churn. The math is brutal: a SaaS at 5% monthly churn cannot grow no matter how much it spends on acquisition. Healthy benchmarks: SMB SaaS 3-5% monthly, mid-market 1-2%, enterprise under 1%. Service retainers should target under 10% annual churn.
Logo churn versus revenue churn versus gross versus net
Four different churn metrics, each tells a different story. Logo churn (customer count): how many customers left. Revenue churn (MRR or ARR): how much recurring revenue left. Gross churn excludes expansion; net churn includes it. A US SaaS with 100 customers at 1000 MRR each (100K total MRR) that loses 5 customers (5K MRR) and grows existing customers by 6K MRR has: 5 percent logo churn, 5 percent gross revenue churn, negative 1 percent net revenue churn (NRR of 101 percent). The four-way breakdown is the standard board-deck format for SaaS. Negative net churn (NRR over 100 percent) is the holy grail because the business grows even with zero new customer acquisition.
Why early-cohort churn is highest
Customers churn at the highest rate in the first 90 days. This is when onboarding gaps, product-market fit mismatches, and overpromised sales messaging surface. US SaaS data shows that 30 to 60 percent of total lifetime churn happens in the first 6 months. The implication: focus customer success heavily on the first 90 days. A user activation milestone (the action that proves value, like 'sent 3 invoices' or 'invited 2 teammates') in the first week predicts retention better than any other signal. Cohort churn curves should flatten by month 6; if they do not, you have a product or fit problem, not a marketing problem.
Causes of churn and their fixes
Five common causes. One, no clear value moment - user signs up, never reaches the aha, drifts away. Fix: redesign onboarding around the value moment. Two, wrong-fit customer acquired by misleading sales messaging. Fix: tighten ICP and qualification. Three, missing features that competitor has - common after first product-market fit when competitors emerge. Fix: ruthless prioritization based on closed-won versus closed-lost interviews. Four, price increase pushed too aggressively. Fix: grandfather long-term customers, communicate value before raising. Five, change in customer's business (downsize, acquired, pivot). Fix: this is unavoidable. Accept 1 to 2 percent monthly churn as floor.
Acceptable churn by segment
US SaaS benchmarks. SMB self-serve: 3 to 7 percent monthly logo churn is the realistic range. SMB sales-assisted: 2 to 4 percent monthly. Mid-market: 1 to 2 percent monthly, or 12 to 20 percent annual. Enterprise: 0.5 to 1 percent monthly, often quoted as 5 to 10 percent annual logo churn with negative net revenue churn. For services businesses with retainers: under 10 percent annual is healthy, under 5 percent annual is exceptional. Project-based services do not have meaningful churn metrics - measure repeat purchase rate instead.
FAQ
What is the difference between monthly and annual churn?
Monthly churn compounds. 5 percent monthly churn does not equal 60 percent annual churn (12 x 5); it equals roughly 46 percent annual churn (1 minus 0.95^12). Conversely, 20 percent annual churn equals roughly 1.8 percent monthly. For SaaS, always state both and clarify which you mean - investors and operators sometimes confuse the two and arrive at wrong LTV calculations.
Is churn a marketing problem or a product problem?
Both, but the diagnosis differs by where churn occurs. Early-cohort churn (first 30 to 90 days) is usually a marketing/sales-fit problem: you acquired customers who do not match ICP. Mid-cohort churn (3 to 12 months) is usually a product problem: customers tried the product, did not get enough value, left. Late-cohort churn (12+ months) is usually a competitive or business-change problem. Run the cohort analysis to know which one you have before assigning the fix.
Should I worry about voluntary versus involuntary churn?
Yes, separately. Voluntary churn is the customer chose to leave; this measures product-market fit and customer satisfaction. Involuntary churn is the payment failed (expired card, insufficient funds); this is operational and fixable with dunning emails, card updaters, and grace periods. US SaaS businesses commonly run 0.5 to 2 percent monthly involuntary churn that disappears with proper dunning. Tools like Stripe Smart Retries, Recurly, and Chargebee automate this.
How do I reduce churn without spending money?
Three high-leverage moves. One, run a quarterly check-in with at-risk customers (low usage, fewer logins) - this alone recovers 10 to 30 percent of would-be churn. Two, improve onboarding so first-week activation rate rises - the single biggest churn predictor. Three, ask churned customers exactly why they left and feed it back to product. Each of these costs founder or CS time, not capital.
What about negative churn?
Negative net churn (net revenue retention above 100 percent) means existing customers contribute more revenue each period through upsells, expansion, and seat additions than is lost through downgrades and cancellations. This is the strongest possible position because the business grows even without new customer acquisition. Best US SaaS companies (Snowflake, Datadog) run 130 to 160 percent NRR. For most service businesses, the equivalent is account expansion - growing retainer value at existing clients faster than losing clients.
In your business
- →Track monthly churn for subscriptions, annual churn for retainers
- →Segment churn by customer cohort - new customers often churn faster than mature ones
- →Exit-interview every churn - patterns emerge fast