L
Glossary
Logo Churn
Logo churn is the percentage of customer accounts that cancel during a period, counted one account at a time regardless of what each one paid. A $400 customer and a $400,000 customer each subtract one logo, so the metric reports how many relationships ended, not what ending them cost.
Key Takeaways
Logo churn weights every account identically, so it catches damage in the small end of the book that dollar-weighted metrics smooth over.
On a 120-account book, 4 losses including one $96,000 account give 3.3% logo churn and 2.47% revenue churn. Next quarter, 9 self-serve losses give 7.6% and 0.92%. One doubles while the other falls by 63%.
CRV puts monthly logo churn near 4.1% in the $500 to $5,000 ACV band, roughly 39% a year, against below 1% monthly above $100,000 ACV.
CRV pairs 5% annual logo churn with 122% net revenue retention in enterprise, so one year reads as shrinking by headcount and growing by dollars.
How do you calculate logo churn?
Divide the accounts that cancelled during the period by the accounts you started with. MetricHQ publishes it as
Nothing in it touches revenue, by design.
Four decisions change the answer before the arithmetic starts:
What counts as one logo. A parent with 6 child accounts is either 1 logo or 6, and the choice moves numerator and denominator.
Whether mid-period signups join the denominator. They dilute the rate, so hold it at the opening count and report new business in MRR movements.
When cancellation lands. Notice given in March for a June contract end churns in June, or quarters stop reconciling.
Whether failed payments count. Involuntary churn fills the logo count with accounts that never chose to leave, so keep it on its own line.
Why does logo churn disagree with revenue churn?
Because one counts accounts and the other counts dollars, and the accounts that leave are rarely average-sized. Here's one book across two quarters:
Quarter | Opening accounts | Opening ARR | Accounts lost | ARR lost | Logo churn | Revenue churn |
|---|---|---|---|---|---|---|
Q2 | 120 | $6,000,000 | 4, one at $96,000 | $148,000 | 3.3% | 2.47% |
Q3 | 118 | $6,300,000 | 9, all self-serve | $58,000 | 7.6% | 0.92% |
Logo churn more than doubled while revenue churn fell by roughly 63%. The average churned account dropped from $37,000 to $6,444, so Q3 lost twice as many relationships for a third of the money.
Read either alone and you get a different quarter. The revenue view says retention improved. The account view says nine teams left, a fit signal that reaches the dollars two or three quarters later. I report both plus the average size of what churned, since that third number explains the gap.
What counts as a healthy logo churn rate?
It depends on contract size, and a benchmark quoted without a segment is unusable. CRV's data and MetricHQ's bands line up closely:
Segment | Monthly logo churn | Annual equivalent |
|---|---|---|
SMB, $500 to $5,000 ACV | around 4.1% | roughly 39% |
Mid-market, healthy | under 2% | under 22% |
Enterprise, above $100,000 ACV | under 1% | roughly 11% or less |
Below $1M ARR, any segment | 3% to 5% | roughly 30% to 45% |
MetricHQ's general rule is annual logo churn under 20%, or under 2% monthly. What moves the bar more than the segment label:
Contract length. MetricHQ puts annual contracts at 10% to 25% a year and multi-year enterprise contracts under 10%, since renewal dates cap how often a customer can leave.
Whether expansion covers the loss. An enterprise cohort losing 5% of its logos a year still posts 122% net revenue retention when growth in surviving accounts outruns the departures, per CRV.
Related terms
The count-based view borders these, and each one answers a question logo churn can't:
Churn rate covers the general formula, denominator choices, and how to make a published benchmark comparable to your own.
Revenue churn is the dollar-weighted twin, the metric logo churn most often contradicts.
Net revenue retention nets expansion against loss, so it clears 100% in quarters where logo churn climbed.
Gross revenue retention ignores every inflow, which makes it the revenue metric closest to what logo churn measures.
Involuntary churn is the share of lost logos caused by a card failing, not a decision.
FAQ
Is logo churn the same as customer churn?
Yes. MetricHQ lists Logo Churn, Customer Churn, and Account Churn as names for the same count-based metric. "Logo" comes from the customer logo wall, where a departure removes one image regardless of contract size.
How do you count a logo when one company holds several subscriptions?
Pick the billing entity that matches how you sell, then hold it constant. Counting each subscription separately turns one enterprise departure into several churned logos and wrecks your segment averages. Counting at the parent hides the loss of individual business units, so most teams count parents and track cancellations separately.
Should logo churn be measured monthly or annually?
Measure it at whatever interval your contracts renew. A self-serve book supports a monthly rate, but an annual-contract book gives most customers one chance a year to leave, so a monthly figure mostly reports which month the renewals landed in. Multiplying a monthly rate by 12 overstates the year, since the base shrinks each month.
Does a reactivated account get counted again?
Yes, if it cancels a second time. A customer who leaves in March and returns in July counts once in March's churned logos and once in July's reactivations. A November cancellation is a second churn event. Netting that account to zero across the year erases both movements and flatters the cohort.
Back to glossary
