A small monthly churn is much bigger than it looks

A monthly churn rate looks harmless on a dashboard. Compounded over a year, and run through lifetime value, it is one of the most important numbers in the business.

Monthly churn compounds

Annual churn is not twelve times the monthly rate, it is one minus the monthly survival rate to the power of twelve. So:

• 2% monthly revenue churn: about 22% a year (1 - 0.98^12).
• 5% monthly revenue churn: about 46% a year (1 - 0.95^12).

Going from 2% to 5% a month does not add a few points to the annual figure, it doubles it.

Churn sets your LTV

Gross-margin lifetime value is the monthly gross margin per account divided by the monthly churn rate. With the default assumptions of 300 euro average revenue per account and an 80% gross margin, the margin is 240 euro a month, so:

• At 2% monthly churn: LTV is about 12,000 euro (240 / 0.02).
• At 5% monthly churn: LTV is about 4,800 euro (240 / 0.05).

The same product, the same price, and the churn rate alone more than halves what a customer is worth. That is the number that decides how much you can afford to spend to acquire one. Put your own MRR, churn, ARPU and margin in and see the annual churn, GRR, NRR and LTV together.

So how much can you spend to acquire a customer?

That lifetime value sets your acquisition budget. The common healthy window is an LTV:CAC ratio of about 3:1 to 5:1, so you divide the LTV by 3 to 5 to get the most you can afford to spend winning one customer:

• At 2% monthly churn, on the 12,000 euro LTV: roughly 2,400 to 4,000 euro per customer.
• At 5% monthly churn, on the 4,800 euro LTV: roughly 960 to 1,600 euro per customer.

So the higher churn rate does not only halve what a customer is worth, it halves what you are allowed to spend to win one. Below 1:1 you lose money on every signup; far above 5:1 usually means you are under-investing in growth.

The ratio is only half the test. CAC payback, the cost to acquire divided by monthly gross margin, tells you how long that money is tied up: a 4,000 euro CAC against 240 euro of monthly margin takes about 17 months to earn back, while a 1,600 euro CAC pays back in about 7. Under 12 months is comfortable, beyond 18 makes growth expensive to fund. The calculator rates your LTV:CAC and CAC payback against these benchmarks alongside the churn and LTV.

Which churn number? Four that get mixed up

The calculator shows more than one retention figure on purpose: "churn" is at least four different numbers, and the LTV above quietly picked one of them.

• Logo (customer) churn counts accounts lost. Revenue churn counts the money lost. They only match if every account pays the same. Lose your smallest accounts and revenue churn is lower than logo churn; lose one whale and it is higher.
• Gross revenue churn is the revenue lost to cancellations and downgrades alone. It can never drop below zero, so it is the honest worst case.
• Net revenue churn subtracts expansion (upgrades, added seats, more usage) from that loss. If existing customers expand faster than others leave, net churn goes negative and the same book of customers grows with no new signups. Net revenue retention above 100% is what that looks like on the dashboard.

This is why the LTV above is a careful answer, not the only one. Divide gross margin by gross revenue churn and you get the cautious lifetime value; divide by net revenue churn and, when expansion outruns losses, the denominator shrinks or vanishes and lifetime value stretches far higher. Same customers, very different number, so always be clear which churn rate you fed in.

The LTV number also assumes churn never changes

Every lifetime value on this page came out of one move, gross margin divided by a churn rate, and that move hides a second assumption on top of which churn number you fed it: that the rate holds steady for the whole life of a customer. Real customer bases do not behave that way. A cohort churns fastest in its first few months, as the people who were a poor fit leave, then the rate falls and flattens as the survivors settle in. Bill Gurley made exactly this the core of his 2012 critique of the formula: churn is almost never constant, so a single blended rate misreads what a customer is really worth.

The direction of the error is predictable. Because the simple formula applies one rate forever, it overstates lifetime value for young cohorts, whose high early churn has not yet burned off, and understates it for old cohorts, whose survivors are stickier than the blended average. The honest fix is a cohort retention curve: track the revenue each month of signups actually keeps over the following twelve to twenty-four months, and read lifetime value off the real curve instead of a single divide. The inputs are not independent either, so pushing ARPA up by raising the price tends to lift churn at the same time, which is why you cannot improve the formula one variable at a time. Treat gross-margin-over-churn as a fast estimate and an upper bound on a young book, not a promise.