Basic Churn & Retention Calculator

Your monthly report says churn is 4.2% and the board nods approvingly because last quarter it was 4.8%. But 4.2% monthly churn compounds to 40% annual customer loss — four out of ten customers gone every year. The common mistake is reading monthly churn as if it is an annual number. Retention is churn’s mirror: 95.8% monthly retention sounds nearly perfect, yet over 12 months it means only 60% of customers survive. The calculator converts between these views so you see both the period rate and the compounded annual reality.

Churn tells you how fast the bucket leaks. Retention tells you how much water stays in the bucket. Neither tells you whether the bucket is filling faster than it drains — that is net growth, which combines new acquisition with losses. A business can have high churn and still grow if acquisition outpaces it, but that is an expensive treadmill.

Time window. Monthly, quarterly, and annual churn rates are not interchangeable. 5% monthly churn is 46% annual churn, not 60% (because compounding works against you nonlinearly). Always specify the period when reporting and convert explicitly: annual churn = 1 − (1 − monthly churn)¹².

Denominator definition. Do you count customers at the start of the period, at the end, or the average? Start-of-period is most common and avoids double-counting customers acquired and churned in the same month. But if you add 200 customers mid-month and 15 churn by month-end, using start-of-period as the denominator ignores the new cohort’s churn entirely. Define the denominator once and stick with it.

Revenue vs logo churn. Losing 50 small accounts and retaining 5 large ones can produce 10% logo churn but 2% MRR churn. Revenue churn captures the financial impact; logo churn captures the breadth of the problem. Report both — they tell different stories.

Net growth = new customers − churned customers. If you acquire 300 and lose 250, net growth is 50 — but the business is running hard just to stay roughly in place. The “churn treadmill” effect means that as the customer base grows, the absolute number of churned customers grows proportionally even if the rate stays flat. At 5% monthly churn, a 1,000-customer company loses 50/month; a 10,000-customer company loses 500/month.

This is why improving retention has a compounding payoff: every point of churn reduction saves more customers as the base scales. A product fix that cuts monthly churn from 5% to 4% saves 10 customers/month at 1,000 users. At 10,000 users, the same fix saves 100 customers/month. Retention improvements scale; acquisition campaigns do not (marginal CAC tends to rise).

Mixing voluntary and involuntary churn. A customer who cancels because they dislike the product (voluntary) is a completely different signal from a customer whose credit card expired (involuntary). Involuntary churn is a billing-operations problem with known fixes (dunning emails, card updater services). Voluntary churn is a product or value-proposition problem. Blending the two inflates the number and obscures the root cause.

Survivorship bias in cohort averages. If you report average churn across all active customers, the long-tenured loyal segment dilutes the rate. New-cohort churn is often 2–3× the mature-cohort rate. Segmenting by cohort age reveals where the real problem is — usually the first 30–90 days.

Confusing gross and net churn. If 50 customers leave but 20 existing customers expand (upgrade their plan), gross churn is 5% but net revenue churn might be 2%. Net revenue churn can even go negative if expansion revenue exceeds lost revenue — a sign of product-market fit in the retained base.

Negative net churn. When expansion revenue from upgrades, cross-sells, and price increases exceeds revenue lost to cancellations and downgrades, net MRR churn goes negative. This is the holy grail of SaaS metrics because it means revenue grows even without new customers. But it can mask underlying logo churn — if 8% of customers leave each month and the survivors expand, the logo problem is still real.

Seasonal churn spikes. B2C subscription products often see churn spike in January (post-holiday cancellations) and after free-trial windows expire. A monthly churn rate that doubles in one month and halves the next is not a crisis — it is seasonality. Use trailing-12-month churn to smooth these effects before comparing periods.

Very small customer bases. With 40 customers, losing 2 is 5% churn. Losing 3 is 7.5%. The statistical noise is enormous at small N. Do not optimise based on monthly churn when the denominator is below a few hundred — look at cohort retention curves over longer windows instead.

The core formulas for customer and revenue churn metrics:

Scenario: A SaaS product starts the month with 2,400 customers at $85 average MRR. During the month, 108 customers cancel (72 voluntary, 36 involuntary). 45 existing customers upgrade, adding $4,500 in MRR. 190 new customers join.

Logo churn: 108 / 2,400 = 4.5% monthly. Annual logo churn = 1 − 0.955¹² = 42.3%. Average lifespan = 1 / 0.045 = 22 months.

Revenue churn: Lost MRR = 108 × $85 = $9,180. Expansion MRR = $4,500. Net MRR churn = ($9,180 − $4,500) / ($2,400 × $85) = $4,680 / $204,000 = 2.3%. Net revenue churn is roughly half the logo churn because upsells partially offset cancellations.

Net growth: End-of-month customers = 2,400 − 108 + 190 = 2,482. Net gain of 82 customers. But the involuntary churn of 36 is recoverable — implementing a dunning sequence that recovers 60% (22 customers) would drop total churn to 86 and push net growth to 104. That single billing fix is worth more than 20 additional paid acquisitions.