What Is Churn Rate and How to Reduce It?
Churn rate is the percentage of customers who stop using or paying for a product during a specific period of time, like a month or a year. If a company starts the month with 1,000 customers and loses 30 of them by the end of the month, its monthly churn rate is 3%. Churn is one of the most closely watched numbers in any subscription business, because a high churn rate means a company has to keep replacing lost customers just to stay the same size, let alone grow.
Quick facts
- The basic formula: Churn rate = (customers lost during the period ÷ customers at the start of the period) × 100
- Churn is usually measured monthly or annually, and both numbers matter — a small-looking monthly churn rate can add up to a large annual loss.
- Customer churn counts people who leave. Revenue churn counts the money lost, which can be different if the customers who leave are smaller or larger than average.
- A "good" churn rate depends heavily on the type of business — consumer apps often tolerate higher churn than B2B software, where a 5%+ annual churn rate is often already considered a concern.
- Churn is the mirror image of retention rate — if churn is 3%, retention for that period is 97%.
The formula, explained
Churn Rate = (Customers Lost During Period / Customers at Start of Period) × 100
A worked example: A project management software company starts the month with 2,000 paying customers. By the end of the month, 60 of them have canceled.
Churn rate = (60 ÷ 2,000) × 100 = 3%
This means the company is losing 3% of its customer base every single month. That might sound small, but compounded over a year, a steady 3% monthly churn rate means losing roughly a third of the customer base annually — which is why even a small-looking monthly number deserves real attention.
Customer churn vs revenue churn
| Customer churn | Revenue churn | |
|---|---|---|
| What it counts | The number of customers who leave | The amount of revenue lost from customers who leave or downgrade |
| Why it can differ from the other | Ten small customers leaving looks the same as ten large customers leaving, in customer-count terms | Ten large customers leaving is a much bigger financial hit than ten small ones, even though the customer-count churn looks identical |
| When it matters most | Understanding how many relationships you're losing | Understanding the actual financial impact of losing customers |
A company can have a low customer churn rate but a high revenue churn rate, if the customers who are leaving happen to be its highest-paying ones — which is why serious businesses track both numbers, not just one.
Why customers actually churn — the common reasons
- They never reached real value. A customer who signed up but never fully understood or used the product's core benefit was likely to leave from the very start — this is why activation rate and onboarding quality matter so much for reducing churn later.
- A competitor offered something meaningfully better. Sometimes churn isn't about your product being bad — it's about someone else's being better for that specific customer's needs.
- The price no longer matches the value they're getting. If a customer's usage drops but the price stays the same, they start questioning whether it's worth it.
- A poor support experience. One bad, unresolved support interaction is enough to push a borderline customer to leave, especially if it happens at a moment when they were already reconsidering.
- The business itself changed. Sometimes churn has nothing to do with the product — a customer's company downsizes, pivots, or simply no longer has the original need the product solved.
How to actually reduce churn, step by step
- Find out where customers are dropping off, using cohort data. Cohort analysis shows whether churn happens mostly right after signup (an onboarding problem) or much later (a value or pricing problem) — these need very different fixes.
- Talk directly to customers who've already churned. A short exit survey or a few direct interviews with recently churned customers often reveals patterns that internal data alone won't show.
- Fix the biggest, clearest reason first, rather than trying to solve every possible cause of churn at once. If most churn happens in the first two weeks, focus on onboarding before anything else.
- Watch for early warning signs before a customer actually cancels. A drop in usage, unanswered emails, or a support complaint that never got resolved are all signals a customer might be at risk — proactively reaching out at that point is far more effective than only reacting once someone has already decided to leave.
- Re-measure regularly, since fixing one cause of churn often reveals the next biggest one underneath it.
Common mistakes when trying to reduce churn
- Focusing only on the moment someone cancels, like adding a "please don't go" offer at cancellation, instead of fixing whatever caused them to want to leave in the first place. This treats the symptom, not the cause.
- Averaging churn across very different customer segments. A single overall churn number can hide a serious problem in one segment (like new customers) while a healthy segment (like long-term customers) makes the overall number look fine.
- Ignoring revenue churn while only watching customer churn. As shown above, these can tell very different stories — a business celebrating low customer churn might be missing that its highest-value customers are the ones leaving.
- Treating a reduced churn rate as "solved" and moving on. Churn should be watched continuously, since new causes emerge as a product, market, and competitive landscape change over time.
FAQ
What counts as a "good" churn rate? It varies a lot by business type. B2B SaaS companies often aim for under 5-7% annual churn; consumer subscription businesses often tolerate higher churn, since individual customers are lower-value and easier to replace. Comparing your churn rate to your own past performance is usually more useful than comparing to a generic industry number.
Is some churn always normal, even for a great product? Yes. Some churn is unavoidable — customers' needs change, budgets get cut, or companies close entirely, none of which reflect a flaw in the product. The goal isn't zero churn, it's making sure churn happens for reasons outside your control, not because of fixable problems like a confusing onboarding experience.
How is churn rate different from retention rate? They're two sides of the same coin. If churn rate is 3% for a given month, retention rate for that same month is 97% — the two numbers should always add up to 100%. See Retention Rate vs Churn Rate.
Should a startup worry about churn from day one? Yes, even with a small number of customers. Early churn data, even from a handful of customers, often reveals real problems with onboarding or product-market fit — problems that are much cheaper to fix with 50 customers than with 5,000.