Churn rate is the share of customers, or of recurring revenue, that a business loses over a given period. It is one of the central metrics for subscription companies, banks, telecoms operators and any business that depends on repeat customers, because the cost of replacing lost customers often decides whether growth is profitable. A churn rate is always stated for a period, most often a month or a year.
The term covers two related measures. Customer churn, sometimes called logo churn in business-to-business software, counts accounts that leave. Revenue churn measures the recurring revenue lost, including downgrades. Churn is the mirror image of the customer retention rate: if 5% of customers leave in a year, 95% are retained.
Formula and example
The basic customer churn rate is:
Customers who join during the period are left out of both the numerator and the denominator, so that new sales do not hide losses. A company that starts a month with 2,000 subscribers and sees 60 of them cancel has a monthly churn rate of , however many new subscribers it signed in the same month. Some companies divide by the average number of customers over the period instead, which gives slightly different results and should be disclosed.
Gross and net revenue churn
Revenue churn is usually measured on monthly recurring revenue (MRR). Gross revenue churn counts revenue lost from cancellations and from contraction MRR, the downgrades of customers who stay:
Net revenue churn also subtracts expansion revenue from upgrades and extra seats sold to the same customers:
Suppose a company starts the month with MRR of 100,000, loses 4,000 from cancellations and 1,000 from downgrades, and gains 6,000 from upgrades. Gross revenue churn is . Net revenue churn is . A negative figure is known as negative churn, and corresponds to net revenue retention (NRR) of 101%.
Converting monthly and annual churn
Monthly and annual churn cannot be converted by simply multiplying or dividing by 12, because each month's losses come from a smaller base. If the monthly rate is constant, the annual rate is:
Here and are written as decimals. A monthly churn rate of 3% gives , or an annual rate of about 30.6%, not 36%. The calculation assumes a steady rate, which rarely holds exactly, since newer customers usually churn faster than long-standing ones.
Voluntary and involuntary churn
Voluntary churn happens when customers decide to leave, because of price, a competitor, poor fit or reduced need. Involuntary churn happens when a subscription ends because a payment fails, for example on an expired or replaced card, even though the customer meant to stay. In card-funded consumer subscriptions involuntary churn can be a large share of total losses, and it is usually the cheaper kind to reduce.
Benchmarks and interpretation
Acceptable churn depends heavily on the market. Consumer subscriptions with low prices and easy cancellation usually churn faster than business software sold on annual contracts, and small-business customers churn faster than large enterprises, which face higher switching costs. For that reason churn is normally compared within a segment rather than across industries.
Managers and investors also read churn against growth. A business with high churn must replace a large part of its customer base every year before it can grow at all, which raises total spending on acquisition and lengthens the time needed to recover it. In banking and telecoms the same idea is often called attrition, and banks apply it to deposit and card accounts as well as to customers.
Reducing churn
Businesses use a range of measures, often guided by cohort analysis that shows when and where customers leave:
- Payment recovery. Dunning sequences, smart retries of declined payments and a card account updater service reduce involuntary churn.
- Onboarding and engagement. Helping customers reach the product's main value quickly lowers early cancellations.
- Plan design. Annual plans, pauses and cheaper tiers give customers an alternative to cancelling outright.
- Early warning. Usage data and churn prediction models flag at-risk accounts for outreach.
- Win-back. Offers to former customers recover part of the lost base.
Churn feeds directly into customer lifetime value (CLV): at a constant churn rate, the expected lifetime of a customer is roughly one divided by the churn rate, so halving churn roughly doubles expected lifetime. Because definitions vary between companies, published churn figures are only comparable when the formula, period and treatment of new customers are known.