Churn rate

Customer churn and revenue churn

Customer churn counts people or accounts: customers lost in the period ÷ customers at the start of the period. Revenue churn counts money: MRR lost to cancellations and downgrades ÷ MRR at the start of the period. The two can differ a lot. If your smallest customers leave and your largest stay, customer churn looks bad while revenue churn looks fine.

Net revenue churn also subtracts expansion revenue from upgrades. When existing customers upgrade by more than others cancel or downgrade, net revenue churn is negative, which means revenue grows even with no new customers.

Monthly and annual churn

Churn compounds, so you cannot multiply a monthly rate by 12. Annual churn is 1 minus (1 minus monthly churn) to the power of 12. A monthly churn of 3 percent is about 31 percent a year, not 36 percent.

Churn also sets how long an average customer stays: roughly 1 ÷ monthly churn. At 3 percent a month, that is about 33 months. This is why churn sits at the bottom of the customer lifetime value formula.

Reducing churn

Most churn happens early, from customers who never reached the moment the product became useful. Improving activation, the first experience, usually cuts churn more than discounts at cancellation do. Ask everyone who cancels for one reason, group the answers each month and fix the most common one. Failed card payments cause a share of churn on their own, so retrying them and emailing customers before a card expires is a quick win.

Example

A product starts March with 200 customers and $6,000 MRR. During March 8 customers cancel, taking $200 of MRR, and 3 customers downgrade, taking another $60. Customer churn is 8 ÷ 200 = 4 percent. Gross revenue churn is $260 ÷ $6,000, about 4.3 percent. If upgrades add $300, net revenue churn is ($260 minus $300) ÷ $6,000, about minus 0.7 percent.

Questions people ask

What is a good churn rate?

It depends on who you sell to. Products for consumers and very small businesses often lose 3 to 7 percent of customers a month. Products sold to larger companies on annual contracts often lose under 1 percent a month. Compare yourself with products that sell to the same kind of customer.

Do new customers count in the churn calculation?

Not in the simple version. Divide by the customers you had at the start of the period, so customers who joined and left in the same month do not distort it. Track them separately, because they point to an onboarding problem.