Customer churn rate, often called logo churn, is the percentage of customers who cancel or fail to renew during a defined period, counted by account rather than by dollars. Each customer is one "logo" and counts equally, so a departing free-tier startup and a departing seven-figure enterprise both subtract exactly one from the base. The standard calculation is customers lost in the period divided by customers at the start of the period, times one hundred. It is the count-based half of the two-metric SaaS churn pair, sitting opposite revenue churn, which weights the same losses by the recurring dollars attached to them.
The word churn traces to the agitation of a butter churn, an analogy for customers moving continuously in and out of a subscription base. As a tracked KPI it rose to prominence in telecom and cable through the 1990s and 2000s, where high churn exposed the poor return on heavy acquisition spending, and the concept carried into media subscriptions and then SaaS. The "logo" qualifier is more recent SaaS and venture-metrics vocabulary, adopted by metrics writers and investors to keep customer-count churn clearly separate from dollar churn; no single coiner is identifiable.
Today it appears on nearly every SaaS board deck and investor update, usually beside net revenue retention. Product, finance, and customer-success teams watch it as a health signal, and competitive-intelligence teams treat a rival's suspected churn as an opening, since customer loss rarely stays fully hidden.
How logo churn is calculated
The core formula is straightforward: divide the number of customers lost during a period by the number of customers at the start of that period, then multiply by one hundred. A base of 400 customers that loses 12 in a month posts a monthly logo churn of 3 percent. Because it counts accounts, not revenue, contract size never enters the arithmetic, so every cancellation is worth exactly one.
The judgment calls live in the definitions, not the math. Teams have to decide what "lost" means: an outright cancellation, a non-renewal at term end, or a downgrade that keeps the logo but strips most of its value. They also have to fix the period and whether to annualize monthly figures, which for high-churn segments is not a simple multiply-by-twelve because the base compounds. Logo churn is also the complement of logo retention, so a 3 percent monthly churn implies 97 percent monthly retention over the same cohort.
Logo churn vs. revenue churn
Logo churn and revenue churn measure the same departures through different lenses, and they frequently disagree. Logo churn counts customers; revenue churn counts the recurring dollars those customers carried. A company can lose a long tail of small accounts and post an ugly logo churn number while revenue churn barely moves, because the departed logos represented little money. The reverse is just as common: a single enterprise account leaving can spike revenue churn while logo churn stays flat.
That divergence is the point of tracking both. Logo churn is the better read on product-market fit and self-serve health, since it treats every customer as an equal vote on whether the product is worth keeping. Revenue churn is the better read on financial exposure. Watching only one hides a real problem: strong revenue churn can mask a broad base quietly eroding at the low end, which is often where a competitor is winning switchers first.
How competitive-intelligence teams infer a rival's churn
Competitors almost never publish churn, so CI teams reconstruct it from public proxies. Review platforms are the richest source: language on G2 or Capterra like "looking for alternatives," "we did not renew," or "we switched to X," plus G2's reasons-for-switching and win/loss data, points directly at logo loss. Falling review volume, or rating subcategories such as support and ease-of-setup slipping below prior levels, suggests the base is souring before it leaves.
Other signals sit off the review sites. Customer logos and case studies quietly disappearing from a competitor's website, customer-success or support layoffs and hiring freezes, and contract-term or pricing changes that read as retention pressure all hint at churn the rival would rather not disclose. Teams that monitor competitor websites, job postings, and pricing pages continuously can catch these shifts as they happen and time displacement or outbound campaigns against a rival that looks newly vulnerable, the kind of change meertrack-style tracking is built to surface.
Common mistakes and limitations
The most common error is reading logo churn in isolation. On its own it says nothing about revenue exposure or expansion, which is why it belongs next to revenue churn and net revenue retention rather than alone on a slide. A related trap is mixing gross and net definitions: gross churn counts only losses, while net churn nets expansion within the existing base against those losses, and comparing one company's gross figure to another's net figure produces a meaningless gap.
Benchmarks are the other pitfall. A monthly churn that is healthy for SMB or prosumer software would be alarming for enterprise, where much longer contracts and higher switching costs push acceptable rates far lower, so a single "good churn rate" number is misleading without a segment attached. Logo churn also flattens nuance it does not distinguish on its own, such as voluntary cancellations versus involuntary failed-payment losses, or customers who leave and quickly rejoin, unless the team explicitly defines how those cases are handled.