Logo Churn

A software company starts a quarter with 200 paying accounts and ten of those cancel by quarter end, so its cohort-based logo churn is 5%. If the ten were small accounts, revenue churn could be much lower; if they were large, the reverse could occur.

Sage defines logo churn as customers lost divided by customers at the beginning of the period, and Chargebee distinguishes customer and revenue churn, so both make the denominator and customer definition central to a useful metric. Define a logo first, deciding whether it is a billing account, legal customer or corporate group, so that a company with three subscriptions is not silently counted as one in one period and three in another.

Choose a period and label dates and cadence, because monthly, quarterly and annual rates are not directly interchangeable. Freeze the starting cohort and count only customers present at the beginning, since new customers who join and leave within the period need a separate metric.

Define 'lost' clearly, because a cancelled contract, expired trial and overdue invoice can have different statuses, and use one consistent effective date, since an account requesting cancellation today may remain active until contract end. Exclude free trials if the metric is for paying customers, as trial conversion is another question and mixing the two changes the denominator.

Document how pauses, reactivations, mergers and acquisitions are treated, because a seasonal suspension may not equal permanent cancellation, two accounts may combine without a lost commercial relationship, and a newly acquired customer base can distort trends unless like-for-like cohorts are shown. Measure revenue too, because logo churn weights a small and large customer equally while gross revenue churn or retention shows a different financial impact.

Separate gross and net revenue, since expansion among surviving accounts can offset lost revenue in a net measure and net growth should not hide many departures. Avoid vanity denominators, because adding new customers to the denominator while counting only old-customer losses makes the rate look artificially low.

Segment accounts by plan, size, industry, acquisition channel and tenure to reveal where churn is concentrated and avoid meaningless averages. Early losses may indicate setup or fit problems, but test that hypothesis with product use and customer feedback, and remember that reasons in CRM may be incomplete or biased, so ask departing customers respectfully and compare with behavioural data.

Watch contract timing, since annual renewals can cause quarterly spikes, and distinguish nonpayment, which may reflect financial distress, billing errors or involuntary card failure. Review data quality, because duplicate account records, migrations and missing end dates can create fake churn, so reconcile CRM and billing systems.

Consider customer value, since high logo churn among low-fee accounts can still strain support and acquisition spending, and do not presume causality, because a new onboarding process followed by lower churn may coincide with a changed customer mix. Set targets in context, since a seasonal product differs from mission-critical annual software and one universal good churn rate is weak guidance, track exits and acquisition separately because a net customer count can rise despite substantial churn, and remember that logo churn answers how many starting relationships were lost and becomes actionable when paired with revenue impact and a sound view of why they left.