Churn Rate
Churn Rate is not an isolated customer success problem. It's the most direct thermometer of the coherence between the sales promise and the value actually delivered.
“Churn is almost never born in the month the customer cancels. It's born in the week they were sold something they didn't really need.”
Otto GTM ObservatoryWhy it must be read by segment and cohort
An aggregate annual churn rate of 12% can hide 4% on enterprise customers and 28% on SMB customers. These are two different problems, with different causes and solutions: the first likely organizational, the second likely ICP fit.
Analyzing churn by acquisition cohort (by month, by channel, by rep who closed the deal) isolates the variables the aggregate figure hides, and is often the fastest way to find the root cause.
Anti-patterns
Common mistake: calculating monthly churn rate and multiplying by 12 to estimate the annual figure. Compounding makes this estimate systematically lower than the real figure, often by several percentage points.
Second mistake: attributing churn exclusively to product or customer success, without analyzing whether lost customers were sold outside the ICP from the start by the sales team.
Practical Application
A company with 640 active customers at the start of the year and 58 customers lost during the year records a 9.06% annual churn rate. Breaking it down by segment, churn on enterprise customers (ARPA > €20,000) is 2.8%, while on the SMB segment (ARPA < €3,000) it rises to 24.1%: the operational priority isn't the product, it's the inbound qualification criteria.