Customer Retention Rate Calculator — From Start, End and New Customer Counts

Background.

Customer retention rate is the share of the customers you began a period with who were still there at the end of it. The formula that matters is not the one most people reach for. Retention is retained customers divided by the customers who were at risk — and the customers you signed during the period were never at risk, so they belong on neither side of the fraction. That means the calculation is (customers at end − new customers) ÷ customers at start, not customers at end ÷ customers at start.

The difference is not academic. Load this page and the default figures describe a business that went from 1,200 customers to 1,320, signing 180 along the way. Divide 1,320 by 1,200 and you get 110% — a number that is not retention at all, and which cannot be retention, because you cannot keep more customers than you had. The correct calculation is (1,320 − 180) ÷ 1,200 = 95%: this business kept 1,140 of its original customers and lost 60. Both facts are true simultaneously — the customer count grew 10% and 5% of the opening base walked out — and the naive figure blurs them into one flattering number. The gap between the two, reported here as its own output, is exactly your acquisition rate: 180 ÷ 1,200 = 15 percentage points. The faster you grow, the more the wrong formula flatters you.

The formula on this page is the one MASB, the Marketing Accountability Standards Board, publishes in its Common Language Marketing Dictionary, drawn from Farris, Bendle, Pfeifer and Reibstein's Marketing Metrics: "Retention rate is the ratio of the number of retained customers to the number at risk." Datadog states the same exclusion principle for the revenue version in its Form 10-K — its retention figure "excludes ARR from new customers in the current period." The base is the opening cohort in both cases, and nothing that arrives later enters it.

One rule needs stating before you enter numbers, because it changes the answer. A customer who churned in an earlier period and came back during this one was not in the opening base, so they count as a new customer here, not a retained one. Counting win-backs as retained is how retention rates above 100% appear in dashboards, and this calculator rejects that input rather than reporting an impossible figure. If you want a metric that can exceed 100%, you want net revenue retention, which is measured in dollars and adds expansion revenue on top.

Finally, retention compounds across periods rather than adding up. Ninety-three percent monthly retention does not mean 100 − 7 × 12 = 16% a year; it means 0.93 to the twelfth power, or 41.86% — a much worse number than the linear arithmetic suggests, though better than it feels. Treat the annualised output as a floor. Fader, Hardie and colleagues showed that observed cohort retention rises as a cohort ages, because the customers most likely to leave leave first, so projecting a single blended rate forward always under-counts the survivors.

What is customer retention rate calculator?

Customer retention rate (CRR) measures the proportion of an existing customer base that is still active at the end of a defined period. It counts customers, not revenue, and it is calculated by dividing the number of retained customers by the number of customers who were at risk at the start of the period, then multiplying by 100. Retained customers are derived here as the closing customer count minus customers acquired during the period, because new arrivals were never exposed to the possibility of leaving and so cannot be part of a retention measure. Customer churn rate — also called logo churn or customer attrition — is the complement of retention over the same base and period: churn = 100% − retention. Retention rate counts customers irrespective of how much each one spends, which is what separates it from revenue retention metrics: a business can retain 95% of its customers while losing far more or far less than 5% of its revenue, depending on the size of the accounts that leave. Because both the numerator and the denominator exclude mid-period acquisitions, customer retention rate is capped at 100% and cannot exceed it. It is an operational metric with no accounting-standard definition, so the period, the denominator convention and the treatment of win-backs all need stating alongside any figure you publish.

How to use this calculator.

1. 01Choose the period your counts cover. Twelve months is the default and the most common reporting basis for retention; use one month or one quarter if that matches your reporting cycle.
2. 02Enter the number of active customers on day one of the period. This is your at-risk base and it becomes the denominator.
3. 03Enter the number of active customers on the last day of the period, including everyone you signed along the way. Do not subtract anything yourself.
4. 04Enter how many customers you acquired during the period who were still active at period end. Count a returning, previously-churned customer as new — they were not in the opening base.
5. 05Read the retention rate together with the summary line beneath it, which states the counts and shows what the same figures would have read had the new customers been left in.
6. 06Compare the retention rate against the net customer growth figure. They answer different questions and can move in opposite directions; a business can grow its count while its base leaks.
7. 07Use the annualised figure as a floor rather than a forecast, and prefer a directly measured twelve-month rate whenever you have one.

The formula.

Customer retention rate is retained customers over the at-risk base: CRR = (E − N) ÷ S × 100, where S is the customer count at the start of the period, E the count at the end, and N the customers acquired during the period who were still active at period end. The subtraction of N is what makes it a retention measure rather than a growth measure. With the default figures — S = 1,200, E = 1,320, N = 180 — retained customers are 1,320 − 180 = 1,140, so CRR = 1,140 ÷ 1,200 = 95.00% and 60 customers were lost. Customer churn is the complement over the same base and period: 100% − 95.00% = 5.00%. Two further outputs exist purely to make the standard error visible. The naive figure E ÷ S = 1,320 ÷ 1,200 = 110.00% is reported and labelled as not being retention, and the gap between it and the real rate is reported as overstatement in percentage points: 110.00 − 95.00 = 15.00 points. That gap is not arbitrary. Algebraically, E ÷ S − (E − N) ÷ S = N ÷ S, so the size of the error is precisely your acquisition rate — 180 ÷ 1,200 = 15.00%. A company acquiring nothing sees the two figures agree; a company doubling its customer base sees the naive figure overstate retention by a hundred points. Net customer growth, (E − S) ÷ S, is reported separately because it is the question the naive formula was accidentally answering: 120 ÷ 1,200 = 10.00%. Annualisation compounds the retention ratio rather than scaling the churn rate: CRR raised to the power 12 ÷ L, where L is the period length in months. All arithmetic is carried at full arbitrary-precision decimal width, and nothing is rounded until the result is returned — the retention ratio in particular is never rounded before being raised to the 12/L power.

A worked example.

Frequently asked questions.

References& sources.

1. [1]MASB (Marketing Accountability Standards Board) — Common Language / Universal Marketing Dictionary, entry "Retention Rate": "Retention rate is the ratio of the number of retained customers to the number at risk. Retention rate is used to count customers and track customer activity, irrespective of the number or dollar value of transactions made by each customer." Formula: "Retention rate (%) = [Number of customers retained ÷ Number of customers at risk] x 100". Sourced by MASB from Farris, P. W., Bendle, N. T., Pfeifer, P. E. & Reibstein, D. J., Marketing Metrics: The Definitive Guide to Measuring Marketing Performance, 2nd ed. (Pearson, 2010). Standards-body reference; free. Retrieved 2026-07-29.
2. [2]Datadog, Inc. — Form 10-K for the fiscal year ended December 31, 2020, Item 7 MD&A: "Current Period ARR includes any expansion and is net of contraction or attrition over the last 12 months, but excludes ARR from new customers in the current period." Cited as an SEC-filed statement of the mid-period-acquisition exclusion principle. Datadog states it for revenue; this page applies the same principle to customer counts. Retrieved 2026-07-29.
3. [3]Banzai International, Inc. — Form 10-K for the fiscal year ended December 31, 2024, Item 7 MD&A, "Customer Churn %": "Customer Churn % is the rate of customers who deactivate in a given period relative to the number of active customers at the beginning of such period or end of the prior period." Cited for the beginning-of-period denominator and the churn/retention complement. Retrieved 2026-07-29.
4. [4]Fader, P. S., Hardie, B. G. S., Liu, Y., Davin, J. & Steenburgh, T. (2018). "'How to Project Customer Retention' Revisited: The Role of Duration Dependence." Journal of Interactive Marketing, 43, 1–16. "Cohort-level retention rates increase because those customers with high churn propensities drop out early on, leaving an ever-increasing proportion of customers who have low propensities to churn." Link is the authors' own full-text copy hosted at London Business School; the Elsevier journal version is paywalled. Retrieved 2026-07-29.
5. [5]U.S. Securities and Exchange Commission — Commission Guidance on Management's Discussion and Analysis of Financial Condition and Results of Operations, Release Nos. 33-10751; 34-88094; FR-87, 85 Fed. Reg. 10568 (Feb. 25, 2020), Section II "Key Performance Indicators and Metrics", pp. 10569–10570: a registrant presenting a metric should accompany it with "a clear definition of the metric and how it is calculated". Official Federal Register text via GPO govinfo. Retrieved 2026-07-29.