Churn rate is the percentage of your customers who stop buying from you, or cancel, during a set period. If a meal-kit service starts June with 400 subscribers and 12 of them cancel during the month, its monthly churn rate is 3%.
How churn rate works
The basic formula is customers lost during the period divided by customers at the start of the period, multiplied by 100. Customers who joined during the period are left out of the calculation, otherwise a busy month for sign-ups hides the people walking out of the back door.
For a subscription or contract business, “lost” is easy to define: the customer cancelled or did not renew. For everyone else you have to set a rule. A shop selling dog food might say a customer has churned if they have not ordered for four months, because most people reorder every five or six weeks. A boiler servicing firm might use 15 months, because the natural cycle is a year. The rule should follow how often a loyal customer normally buys.
There are two versions worth knowing. Customer churn counts heads. Revenue churn counts money, so losing one large account can matter more than losing ten small ones. In B2B and software, upgrades from the customers who stay can outweigh what the leavers took with them, which is why some firms report net revenue churn below zero.
Be careful converting between periods. A 3% monthly churn rate is not 36% a year. Each month you keep 97% of the customers left, so after twelve months you keep about 69%, an annual churn rate of roughly 31%.
Why it matters
Churn puts a ceiling on growth. If you win 30 new customers a month and lose 30, you are working hard to stand still. It also sets how long the average customer stays: at 3% monthly churn the average relationship lasts about 33 months (one divided by the churn rate), and that figure feeds straight into customer lifetime value, which in turn decides how much you can afford to pay in advertising to win a customer.
For UK subscription businesses there is a legal angle too. The subscription contract rules in the Digital Markets, Competition and Consumers Act 2024 are built around clear reminders and easy cancellation, so keeping customers by making it hard to leave is not a plan. Check the current commencement position of those rules when you read this, but design for them now.
Common mistakes
- Counting new customers in the starting figure. It flatters the rate in growth months.
- Using one rate for everyone. New customers usually leave faster than those who have stayed a year, so a single average hides where the problem is. Cohort analysis shows it.
- Treating failed payments as choices. Expired cards and failed direct debits cause involuntary churn, which is often fixable with reminders and retry rules.
- Watching acquisition only. A marketing report full of new sign-ups can look healthy while the customer base quietly shrinks.
How to act on it
Calculate churn every month with a fixed definition, then split it by the month customers joined and by the channel that brought them in. If customers from one ad campaign leave twice as fast as those from organic search, that campaign is more expensive than its cost per sale suggests.
Next, separate voluntary from involuntary churn and fix the involuntary kind first, since it is usually the cheapest win. Add a one-question exit survey to the cancellation flow and read the answers every month. Look hard at the first 30 to 60 days, when most avoidable churn happens: onboarding emails, a check-in call or a simple “how is it going” message often matter more than any discount.
Finally, put churn into your targets. Track retention rate alongside it, and set acquisition budgets from lifetime value rather than first-order revenue. Building those numbers into one plan is part of the digital marketing strategy work I do with clients.