SaaS Churn Rate Calculator
Separate logo loss from recurring-revenue erosion, preserve the opening cohort denominator, and see what repeated monthly churn could do to a subscription base over twelve comparable periods.
Build the monthly cohort bridge
Formulas that keep the denominator honest
Logo churn = lost opening customers / customers at start
Logo retention = 1 - logo churn
Gross MRR churn = (canceled MRR + contraction MRR) / starting MRR
NRR = (starting MRR - canceled MRR - contraction MRR + expansion MRR) / starting MRR
Comparable 12-month survival = (one-month retention rate) raised to the 12th power
The opening base is the denominator because the question is what happened to the customers or revenue present when the measurement period began. Putting new customers into the denominator can make acquisition mask retention weakness. New revenue still matters; it belongs in a separate gross-new or net-change bridge.
Worked monthly subscription example
The default company starts with 1,000 paying customers. Forty of those customers cancel, so logo churn is 4.00 percent and opening-logo retention is 96.00 percent. It also adds 120 new customers, giving an ending count of 1,080 and net customer growth of 8.00 percent. That growth does not change the 4.00 percent loss rate among the starting logos.
The opening cohort produces $100,000 of monthly recurring revenue. Full cancellations remove $5,000 and downgrades remove another $3,000. Gross MRR churn is therefore 8.00 percent, leaving $92,000 before expansion. Existing customers expand by $4,000, making ending MRR from that opening cohort $96,000 and NRR 96.00 percent. New-customer MRR is intentionally absent because NRR follows only the opening cohort.
If the monthly logo retention stayed exactly 96 percent for twelve comparable months, about 61.27 percent of the original logos would remain. If gross revenue retention stayed 92 percent each month, only about 36.77 percent of opening MRR would survive before expansion. Repeating 96 percent NRR yields a 61.27 percent cohort-revenue path. These are compounding illustrations, not forecasts: seasonality, contract terms, cohort age, pricing, and expansion opportunities normally change over time.
Logo churn and revenue churn are not substitutes
Logo churn
Counts customers. Losing one small account and one enterprise account produces two lost logos even though the revenue effect may differ dramatically.
Gross MRR churn
Measures canceled and contracted recurring revenue before expansion. It reveals the leakage that upsell cannot erase operationally.
Net revenue retention
Adds expansion from the opening cohort. NRR can exceed 100 percent even while some customers leave, so it must not be labeled customer retention.
A board packet should state which metric is being discussed. “Churn improved” is incomplete without the unit, cohort, period, inclusion rules, and whether expansion is netted against losses. Keep a glossary beside the dashboard so sales, customer success, finance, and investors use the same language.
Define the customer and the opening cohort
Decide whether a customer is an account, parent organization, workspace, paid seat, subscription, location, or contract. A customer with three products can cancel one module while remaining an active logo. A merger of duplicate accounts can look like churn even when no commercial relationship was lost. Free, trial, paused, delinquent, and usage-only accounts require explicit treatment.
Freeze the opening cohort at the first moment of the period, then follow its events. If an opening customer churns and returns in the same month, the documented policy must say whether that is a churn and reactivation or uninterrupted service. Backdated cancellations and late billing corrections should be assigned consistently, with prior-period restatements disclosed when material.
Use identifiers that persist through billing-system migrations and corporate hierarchy changes. Reconcile the count to the billing subledger and the revenue bridge. If finance recognizes revenue on a different timetable from billing MRR, label the operational metric as MRR rather than GAAP revenue.
Monthly, quarterly, and annual churn require compounding
Multiplying a monthly churn percentage by twelve is only a rough approximation and can exceed logical bounds. Survival compounds because each later loss applies to the smaller cohort that remains. At a constant monthly retention rate of 96 percent, annual survival is 0.96 raised to the twelfth power, not 52 percent from simply subtracting 48 percentage points.
Quarterly metrics should be built directly from a quarterly opening cohort when possible. Averaging three monthly churn percentages gives equal weight to months with different opening bases. A weighted bridge using total eligible opening exposure can be more informative, but it is a different construction that should be documented.
Annualizing one month is especially fragile for seasonal products, annual renewals, implementation cycles, price changes, or small account bases. Show the actual rolling twelve-month cohort alongside any run rate. A confidence interval or raw count can prevent a one-customer move from being overinterpreted.
Build an MRR bridge that can be audited
Start with recurring subscription value at a fixed point in time. Remove full cancellations and contraction from the same opening customers, then add expansion from those customers. Keep new-logo MRR separate. The bridge should reconcile to ending recurring revenue after allowing for reactivations, currency, acquisition, divestiture, and classification changes.
Exclude one-time implementation, hardware, services, taxes collected, pass-through usage, and other nonrecurring amounts unless the formal policy treats them as recurring. Usage-based revenue needs a normalization rule because ordinary volume volatility can look like expansion or contraction. Annual contracts may be divided by twelve for MRR, but discounts, ramps, renewal changes, and contract modifications still need a consistent rule.
Do not silently offset bad debt or involuntary delinquency against voluntary churn. Both affect economics, yet different teams own the remedies. Tag churn reason, product, plan, tenure, acquisition channel, geography, contract term, and account segment so the rate leads to action.
Turn churn into a retention operating system
Segment first. Early-life churn may indicate poor acquisition fit or onboarding. Renewal churn may point to weak realized value, procurement friction, competitive pressure, or budget cycles. Contraction can signal unused seats, product gaps, price packaging, or deliberate customer optimization. Involuntary churn may be addressed through payment retries and card updates rather than product changes.
Track leading signals such as activation, time to first value, depth and frequency of use, administrator engagement, support severity, invoice status, champion turnover, integration health, and executive-business-review outcomes. Then test interventions against comparable control groups. A discount that delays cancellation can reduce churn while destroying lifetime contribution if the price concession is not measured.
Pair retention with acquisition efficiency. A company can grow customer count while replacing a weak cohort with expensive new logos. Conversely, deliberate pruning of unprofitable accounts can raise apparent churn while improving cash generation. Review gross margin, customer-acquisition cost, payback, support cost, and lifetime value with the churn bridge.
Public-company definitions show why labels matter
SEC filings illustrate that issuers use company-specific definitions. One filing defines customer churn using deactivations divided by beginning active customers and distinguishes logo from revenue churn. Another describes online average monthly churn using entry MRR and quarterly MRR churn. Those examples are useful evidence that a label alone does not guarantee comparability.
When presenting a non-GAAP or operating KPI, explain why management uses it, how it is calculated, and any change in method. Keep source data, transformation logic, and manual adjustments. If a metric is material to investors, involve accounting, legal, controls, and investor-relations teams rather than letting dashboard code become the policy.
The calculator produces internal analytical outputs. It does not determine securities-law disclosure, GAAP revenue, deferred revenue, bookings, remaining performance obligations, or tax.
Monthly churn-close checklist
- Freeze the opening logo list.
- Reconcile opening MRR to billing.
- Apply the written customer definition.
- Separate cancellation from contraction.
- Keep new logos outside churn denominators.
- Tag reactivations consistently.
- Normalize currency under policy.
- Review mergers and duplicate accounts.
- Exclude nonrecurring revenue consistently.
- Reconcile the ending bridge.
- Segment by cohort and plan.
- Document manual corrections.
Frequently asked questions
Should new customers reduce the churn rate?
No. Churn evaluates losses from the opening population. New customers affect net growth and ending count, which should be shown separately.
Can NRR be above 100 percent?
Yes. Expansion from retained opening customers can exceed cancellations and contraction. That does not mean logo churn is zero.
Should contraction count as customer churn?
Usually not for logo churn because the account remains active. It does belong in gross recurring-revenue churn under the definition used here.
Is monthly churn times twelve annual churn?
No. Retention compounds. Use one minus the monthly churn rate, raise that retention rate to the twelfth power, then subtract from one for an illustrative annual churn rate.
What is a good SaaS churn rate?
There is no universal threshold. Contract term, market, account size, product maturity, pricing, gross margin, expansion, and acquisition cost determine what is sustainable.
Once churn has been measured consistently, carry that retention assumption into the SaaS valuation calculator to examine its effect on enterprise value.