Churn rate is the floor underneath every retention and growth metric. If the floor leaks, no amount of new ARR fills the hole quickly enough, which is why boards read churn before they read growth.
Churn rate is the percentage of customers or recurring revenue a subscription business loses in a defined period. Logo churn counts the share of customer accounts that cancelled. Dollar churn counts the share of recurring revenue that walked out the door. Both are reported monthly or annually. Best-in-class B2B SaaS targets under 1% monthly logo churn and under 5% annual gross dollar churn, and uses the number as the primary input to net and gross revenue retention.
The six things to understand about churn rate before you report it to a board, benchmark it against the market, or build a retention playbook around it. Each one is a place real operators lose credibility by picking the wrong definition or the wrong window.
Churn rate is the share of customers or recurring revenue a subscription business lost during a defined window. It is always a percentage, always tied to a specific period (monthly or annual), and always calculated against the base that existed at the start of that period. The number without a window attached is not a churn rate, it is a slogan.
Logo churn counts customer accounts that cancelled, regardless of their size. Dollar churn counts the recurring revenue those accounts carried. The two numbers can diverge sharply. Losing many small accounts drives logo churn without moving dollar churn. Losing one large account does the opposite. Honest reporting shows both, side by side, every period.
Monthly churn is the right view for high-velocity SMB books where customers cancel month-to-month. Annual churn is the right view for enterprise books where contracts are twelve months or longer. Mixing the two misleads: a 2% monthly churn rate is roughly 22% annualized, not 2% annual, and treating them as the same number is a frequent reporting error.
Best-in-class B2B SaaS targets monthly logo churn below 1% and annual gross dollar churn below 5%. SMB books run higher by segment nature, and enterprise books run lower. A business materially outside these guideposts usually has a product-market fit problem, a target-segment problem, or an onboarding problem rather than a renewal-tactics problem.
Churn rate is the primary input to gross revenue retention (GRR) and net revenue retention (NRR). GRR is one minus gross dollar churn. NRR is GRR plus expansion from the same cohort. A business cannot credibly report retention metrics without first publishing the churn rate those numbers were derived from, and boards expect to see the two together.
The churn number is the signal that triggers the retention playbook. Rising churn in a cohort or segment is the earliest leading indicator that something upstream, in product fit, onboarding, support, or pricing, is broken. Operators who treat the number as a lagging board metric miss the window to intervene. Operators who treat it as a trigger act on it in the quarter it moves.
Logo churn and dollar churn answer different questions about the same cancellation event. Logo churn asks how many accounts walked. Dollar churn asks how much recurring revenue walked. The six cards below walk through what each one measures, where they diverge, and why mature operators always report both.
Logo churn is the number of customer accounts that cancelled in a period, divided by the number of customer accounts active at the start of the period. It treats every account equally, regardless of contract size. A $500 per month account and a $50,000 per month account each count as one logo in the calculation.
Dollar churn is the recurring revenue from cancelled accounts in a period, divided by the recurring revenue of all accounts active at the start of the period. It weights every cancellation by its contract size. One large enterprise cancellation can move dollar churn by multiple points while barely registering on logo churn.
When logo churn is high and dollar churn is low, the business is losing small accounts and keeping large ones, usually a signal that a product fits enterprise better than SMB. When dollar churn is high and logo churn is low, the business is losing a few large accounts, usually a signal of a concentration risk that one bad quarter can concentrate further.
Gross dollar churn counts only full cancellations. Net dollar churn also includes downgrades (customers who stayed but shrank). A customer who dropped from a $5,000 plan to a $2,000 plan contributes $3,000 to net churn but zero to gross churn. Reporting one as the other, without the label, is a recurring source of overstated retention.
A customer whose card declined and whose subscription auto-cancelled is still churn. Mature teams separate voluntary churn (the customer chose to cancel) from involuntary churn (payment failure) because the playbook for each is completely different. Involuntary churn is addressed with dunning and card-updating, voluntary churn with product and success.
The convention honest teams follow is to publish logo churn and dollar churn in the same board slide, with the same window label, every period. Showing only one of them hides the shape of the business. Showing both forces an answer to the question of whether the churn is a size-of-customer problem or a count-of-customer problem.
Churn rate is one of the simplest formulas in SaaS, which is also why it is one of the most frequently miscalculated. The six cards below walk through the standard formulas, the window conventions, and the three recurring errors that make a reported churn number fall apart under board scrutiny.
Logo churn rate equals the number of customer accounts that cancelled in the period, divided by the number of customer accounts active at the start of the period, expressed as a percentage. New accounts added during the period are excluded from the denominator, which is why starting the denominator from the opening balance is the standard convention.
Gross dollar churn rate equals the recurring revenue from cancelled accounts in the period, divided by the recurring revenue of all accounts active at the start of the period. The numerator is the annualized value of the lost contracts, not their one-period billings. The denominator is the opening ARR balance for the cohort being measured.
Monthly churn is measured over a single calendar month. Starting ARR is the opening balance on day one. Cancellations are everything processed inside the month. This is the right window for SMB subscription books where contracts are month-to-month and cancellations happen on a continuous rolling basis, not clumped around renewal dates.
Annual churn is measured over a trailing twelve-month window. For enterprise books with twelve-month-plus contracts, cancellations clump around renewal dates, so a single month is noise. The annual view smooths that out and reflects the business a renewing contract actually renews against. Reporting monthly for enterprise is almost always misleading.
The most common error is reporting a monthly churn rate times twelve as the annual churn rate. Churn compounds, it does not add. A 2% monthly rate is roughly 1 minus 0.98 to the twelfth, or about 21.5% annualized, not 24%. For back-of-the-envelope work the error is small, but on a board slide it matters.
The second recurring error is using the average or ending account count as the denominator instead of the starting count. Each choice tells a different story and each has a defender, but the convention in SaaS reporting is the starting balance. The important thing is to pick one, document it, and apply it the same way every period.
A churn rate is only useful against a benchmark and a playbook. The six cards below describe the widely referenced benchmarks for B2B SaaS, the retention motions that strong operators run against them, and the CRM shape that produces defensible numbers in the first place. Numbers without the system behind them are a report, not a program.
High-velocity SMB subscription books typically run 1% to 2% monthly logo churn, which annualizes into the 10% to 20% range. Below 1% monthly is best-in-class. Above 3% monthly is a warning sign that the business is winning accounts it cannot keep, usually because of mis-targeted acquisition or shallow product fit for the segment it is selling to.
Enterprise B2B SaaS books target annual gross dollar churn below 5%, with best-in-class at or below 2%. The low number reflects that enterprise contracts are long, deeply integrated, and expensive to switch. Enterprise annual churn materially above 10% is almost always a signal of a product that was oversold into a segment it was not ready to serve.
Churn rate is a leading indicator, not a lagging one. Rising churn in a recent cohort shows up quarters before it dominates the aggregate. Honest teams report churn by cohort vintage, by segment, and by acquisition channel precisely so a bad new cohort is caught before it accumulates. Reading only the aggregate rate buys silence until it is too late.
When a cohort shows rising churn, the retention playbook kicks in: proactive renewal outreach, executive sponsorship on at-risk accounts, product usage reviews, pricing adjustments, and in some cases a product fix that the churn signal just forced into visibility. The playbook is only useful because the churn number arrived in time to run it.
A credible churn number requires that every cancellation lands on the customer account record in the CRM, with a date, a reason code, a lost ARR amount, and the user who processed it. From that primitive data the churn rate rolls up by segment, by cohort, by product, by rep, and by acquisition source without a spreadsheet in the loop.
The last defensive step is reconciliation: every cancelled subscription in the CRM maps to a cancelled subscription in the billing system, and every stopped invoice in billing maps to a cancellation record in the CRM. Discrepancies are investigated, not papered over. The reconciliation is what makes the churn rate hold up in diligence instead of collapsing under it.
Strkr is a CRM that captures every cancellation on the customer account with a reason code, a lost ARR amount, and a date, and rolls that up into logo churn and dollar churn by segment, cohort, and acquisition source. The retention numbers reconcile to the system the revenue team already uses, instead of being reassembled from spreadsheets at month end.
Logo churn rate equals the number of customer accounts cancelled during the period divided by the number of accounts active at the start of the period, expressed as a percentage. Dollar churn rate equals the recurring revenue lost from cancellations in the period divided by the recurring revenue active at the start of the period. New customers added during the period are excluded from the denominator in both cases.
Best-in-class B2B SaaS targets monthly logo churn below 1% and annual gross dollar churn below 5%. SMB subscription books typically run 1% to 2% monthly, which annualizes into the 10% to 20% range. Enterprise books with twelve-month-plus contracts target annual gross dollar churn below 5%, with best-in-class at or below 2%. Benchmarks shift by segment, stage, and market, so the numbers above are guideposts rather than hard thresholds.
Logo churn counts the share of customer accounts that cancelled in a period, treating every account as equal regardless of size. Dollar churn counts the share of recurring revenue those cancellations carried, weighting each one by its contract size. The two numbers can diverge sharply. Losing many small accounts moves logo churn without moving dollar churn. Losing one large account does the opposite. Honest teams report both.
Monthly churn is the right window for high-velocity SMB subscription books where customers cancel month-to-month on a continuous basis. Annual churn is the right window for enterprise books where contracts are twelve months or longer and cancellations clump around renewal dates. A single monthly rate for enterprise is almost always noise, and a single annual rate for SMB misses too much of the signal. Match the window to the contract shape.
Gross dollar churn counts only full cancellations. Net dollar churn also includes downgrades, where a customer stayed but shrank their contract. A customer who moved from a $5,000 plan to a $2,000 plan contributes $3,000 to net churn and zero to gross churn. Reporting one as the other, without the label, overstates retention. Boards expect gross and net churn to be published side by side.
Gross revenue retention (GRR) equals one minus gross dollar churn, expressed as a percentage of the opening ARR balance. Net revenue retention (NRR) equals GRR plus expansion from the same cohort over the same period. A business cannot credibly report NRR or GRR without first publishing the churn rate those numbers were derived from, because the retention metrics are mechanical derivatives of the churn number.
The most common root causes are weak product-market fit for the segment being sold to, a mismatch between what sales promised and what the product delivers, shallow onboarding that leaves customers failing to realize value, pricing that outruns the perceived outcome, and involuntary churn from failed payments. The churn number itself is a signal. The playbook is upstream in acquisition, onboarding, product, and billing.
Monthly for internal operating cadence, quarterly for board reporting, and at a labelled point in time for any investor conversation. Churn should be reported by cohort vintage, by segment, and by acquisition channel in addition to the aggregate, because a bad new cohort is the earliest signal of a problem and the aggregate hides it for several quarters. Every reported number should carry the date it was taken and the window it covers.
No sales call, no migration consultant, no four-month implementation. Enter your card, get 14 days of the full Pro tier, cancel any time before day 14 with zero charge. Spin up a workspace, import your CSV, and have something useful before lunch.
