Marketing strategy
Churn Rate: What It Is, How to Calculate It and How to Reduce It
The formula fits on one line, yet most people get it wrong. Here you calculate it right, convert it to annual and use it even if you run a store or a service with no subscriptions.
Churn rate is the percentage of customers who cancel or stop buying from you in a given period: start the month with 400 and lose 18, and your churn rate is 4.5%. It measures what leaks out the bottom while you pour new customers in the top.
The pages ranking for it are mostly written by subscription software companies, and almost none of them explains what to do if you run a store or a service business. When we asked ChatGPT what a good rate looks like, it gave us benchmarks with no sources at all.
Here you get both formulas done right, the monthly-to-annual conversion, a way to measure it without subscriptions and the five moves that bring it down the most. It's one of the numbers that matters most in your marketing strategy: acquisition without retention is filling a leaky bucket.
On this page
What is churn rate (and what it isn't)
Churn rate is the metric that tells you what share of your customers leaves during a specific period, usually a month or a year.
Customer churn is the loss itself, the people who leave; customer churn rate is that loss as a percentage. In subscriptions and telecom it's also called the cancellation or attrition rate.
What it isn't: the opposite of acquisition, or something more ads will fix. It's also not your website's bounce rate. Churn is the number that tells you whether your product delivers what the sale promised.
How to calculate churn rate: the two formulas
The churn rate formula is simple: divide the customers you lost during the period by the customers you had at the start, then multiply by 100.
| Churn | Formula | When to use it |
|---|---|---|
| Customer churn | Customers lost ÷ customers at start × 100 | Always, it's the baseline |
| Revenue churn | Recurring revenue lost ÷ recurring revenue at start × 100 | When customers pay different amounts |
-
Pick the period and define an active customer
A month, a quarter or a year. An active customer is one who pays a fee or has bought within your repurchase window.
-
Count active customers on day one
That's your base. Anyone who joins during the period doesn't count here.
-
Count how many of them left
Only customers from the starting base who canceled, didn't renew or didn't buy again.
-
Divide and multiply by 100
Lost divided by starting base, times 100. That's your customer churn rate.
-
Repeat it with revenue
What the lost customers paid each month, divided by your recurring revenue on day one. That's your revenue churn.
-
Convert it to annual
Annual = 1 − (1 − monthly churn) to the power of 12. The percentage calculator does it in a minute.
Here's how to calculate churn rate with a made-up gym for this guide: on March 1 it has 400 members paying $50 a month. In March, 18 cancel, paying $45 on average, and 30 new members join.
- Customer churn: 18 ÷ 400 = 4.5%.
- Revenue churn: $810 lost ÷ $20,000 in monthly fees = 4.1%.
- The wrong formula: (400 − 412) ÷ 400 = −3%. It looks like nobody left, and 18 people did.
What is negative churn?
It's when the customers who stay spend more than the ones who leave stop paying, so revenue churn drops below zero. It happens when you upgrade customers or sell them add-ons: you lose accounts and still bill more from the same base.
What's the difference between voluntary and involuntary churn?
Voluntary churn is a customer deciding to leave. Involuntary churn is a customer you lose by accident, usually a failed payment from an expired or maxed-out card. In the July 2026 data Recurly publishes from its network, about a third of all churn is involuntary.
Monthly churn rate vs. annual churn rate
A monthly churn rate doesn't multiply by 12. It compounds, because each month you lose a share of a base that's already smaller.
| Monthly churn | Equivalent annual churn rate | Average customer lifetime |
|---|---|---|
| 1% | 11.4% | 100 months |
| 2% | 21.5% | 50 months |
| 3% | 30.6% | 33 months |
| 5% | 46.0% | 20 months |
| 8% | 63.2% | 12.5 months |
| 10% | 71.8% | 10 months |
We calculated this table with the formula above.
How long does the average customer stay?
Average lifetime is 1 divided by monthly churn: at 5%, a customer stays 20 months on average. That number feeds straight into customer value, which is why a single point of churn is worth so much.
What is a good churn rate? (And why benchmarks mislead)
A good churn rate is one that keeps falling against your own history; the benchmarks floating around contradict each other and rarely say how they were measured.
We tested it: we asked ChatGPT and Perplexity in the US, Spain and Brazil what a good churn rate is for businesses that aren't SaaS, like gyms, agencies and stores. For a gym, ChatGPT said 5% to 10% a month. Perplexity said 3% to 5% in the US and 3% to 7% in Brazil. ChatGPT's US answer cited no sources at all.
Even Recurly's benchmark page, one of the most quoted, gives two different numbers for SaaS (3.04% and 3.22%). Treat benchmarks as a rough guide, never as a target: compare the same definition, the same period and the same kind of customer.
How to measure customer churn without subscriptions
Without subscriptions, you measure churn with a repurchase window: a customer counts as lost if they don't buy again in the time they normally would.
- Work out how often repeat customers come back. That's your repurchase cycle.
- Set a window of twice that cycle: if they repeat every 45 days, a customer has churned after 90 days without an order.
- Measure by cohort: of the people who bought in January, how many haven't come back within that window?
It works the same for an online store, a hair salon or a bookkeeping firm. If your catalog is made of digital products to sell once, like courses or templates, track how many buyers come back for a second product instead: that's your real customer retention rate.
Cohort analysis for churn: how to read retention by group
A cohort is a group of customers who share something, usually the month they signed up. Cohort analysis tracks each group month by month, so you can see whether newer customers stay longer or leave sooner than older ones.
A spreadsheet is enough: one row per cohort and one column per month since signup. Pull each month's signups from your billing system or order history, then count how many are still active one, two and three months later. An invented example, with the gym from earlier:
| Cohort | Joined | After month 1 | After month 2 | After month 3 |
|---|---|---|---|---|
| January | 100 | 92% | 87% | 84% |
| February (free enrollment) | 120 | 82% | 70% | 64% |
Read across the rows, not down the columns. February loses 18% in its first month against 8% for January: the problem isn't the gym, it's who the free-enrollment promo attracts. In both rows the biggest drop comes early, so that's where a welcome call belongs.
Always compare cohorts of the same age: a three-month-old customer with another three-month-old one, never with a one-year veteran.
Churn and customer lifetime value: what keeping a customer is worth
Customer lifetime value is what a customer leaves you over the whole relationship, and it depends directly on churn.
The quick version: monthly fee × margin ÷ monthly churn. For the gym in the example, $50 × 60% ÷ 4.5% = $667 per member. Cut churn to 3.5% and each member is worth $857, 29% more, without signing up anyone new.
Research by Frederick Reichheld of Bain & Company, cited in Harvard Business Review, found that raising retention by 5% can lift profits by 25% to 95%. Run your own numbers with the percentage calculator.
Net revenue retention (NRR): how to calculate it and read it
Net revenue retention shows how much of last period's recurring revenue you keep from the same customers, after cancellations, downgrades and upgrades. New customers don't count. Above 100%, the customers you kept grew faster than the ones you lost.
The formula: (starting recurring revenue + upgrades − downgrades − cancellations) ÷ starting recurring revenue × 100.
Take the gym again, with $20,000 in monthly fees on March 1. During March, 25 members move up from $50 to $65 (+$375), 10 move down to $40 (−$100) and 18 cancel (−$810).
- Net revenue retention: (20,000 + 375 − 100 − 810) ÷ 20,000 = 97.3%.
- What it takes to pass 100%: upgrades above $910, about 61 members moving up by $15.
NRR is the number behind negative churn: net revenue churn is simply 100% minus NRR. It also explains why the two churn rates can disagree. A gym can lose 4.5% of its members and still sit above 100% if the rest upgrade. Without plans or add-ons to upgrade to, NRR is just 100% minus revenue churn. And since it only counts the customers you started the month with, a big month of new sign-ups can't flatter it.
How to reduce churn: 5 moves that actually change the number
To reduce churn, go after the cancellations nobody wanted first, then fix how customers start: those are the two biggest leaks and the cheapest to plug.
- Fix failed payments. Warn customers before their card expires, retry the charge on different days and offer a second payment method. That's where involuntary churn lives.
- Nail the first 30 days. Get customers to their first result in week one: a call, a short guide or a guided first session.
- Watch the signals. Less usage, fewer visits, fewer orders: reach out before they ask to cancel, not after.
- Offer a pause instead of a cancellation. And always ask why they're leaving: three repeated reasons are your to-do list.
- Reward commitment. An annual plan with a discount or a loyalty program stretches the relationship.
Measure the effect with the same formula every month. Better targeting helps too: a high conversion rate full of customers who don't fit gets paid back in churn three months later. And once churn is under control, show it off: low churn is one of the most convincing numbers in a pitch deck.
Act as a retention analyst. Here is my data for the last 12 months, month by month: customers at the start, cancellations and new sign-ups [paste the table], plus the average monthly fee [number]. 1) Calculate my monthly customer churn and revenue churn without mixing cancellations and sign-ups. 2) Convert it to annual and calculate average customer lifetime. 3) Tell me which months spike and what could explain them. 4) Suggest three actions ranked by impact and cost. Don't invent data I haven't given you.
Frequently asked questions about churn rate
What is employee churn rate?
It's the same idea applied to your team: the share of employees who leave in a given period. You calculate it the same way, departures during the period divided by headcount at the start, and HR usually calls it turnover or attrition. Split voluntary exits from layoffs, because a high voluntary rate tends to flag problems with pay, management or workload before they show up anywhere else.
What does churn rate mean in sales?
In sales, churn usually means accounts that don't renew or stop ordering. Sales teams often track it two ways: logo churn, the number of accounts lost, and net revenue retention, which compares this year's revenue from existing accounts with last year's, upgrades included. A team can lose accounts and still grow if the accounts that stay spend more.
What is the opposite of churn rate?
Customer retention rate: the share of the customers you started with who are still with you at the end. If you lose 4% of your customers in a month, you retain 96%. Both use the same starting base and period, so they always add up to 100. Many stores prefer to report retention because it's the number they want to see go up.
Is 90% customer retention good?
It depends on the period. A 90% retention rate per year means 10% annual churn, a figure many subscription businesses would be happy with. A 90% retention rate per month means 10% monthly churn, which compounds to about 72% a year, so you would lose most of your customers within twelve months. Always check whether the number is monthly or annual before judging it, and compare it with your own history first.
What is a good LTV to CAC ratio?
CAC is what it costs to win a customer and LTV is what that customer leaves you over the whole relationship. A common rule of thumb in subscription businesses is around 3 to 1. Below 1 to 1 you lose money on every customer you add. Churn drives the ratio, because the sooner customers leave, the lower their lifetime value.
Can NPS predict churn?
It helps, but it isn't enough. NPS asks how likely customers are to recommend you on a 0 to 10 scale, and detractors, who answer 0 to 6, tend to leave more often than promoters. But satisfied customers still leave over price or because they no longer need you. Use NPS as one signal next to usage and payments, and call every detractor the same week.
What is MRR and why does it matter for churn?
MRR is monthly recurring revenue: what you bill every month from subscriptions or fixed fees. It matters because customers aren't all worth the same. Losing five small accounts can hurt less than losing one big one, and only revenue churn, calculated on MRR, shows you that. If you have different price tiers, always track both churn rates side by side.