Churn Rate, Explained: The Formula, Customer vs Revenue, and Why Small Numbers Hurt

Churn is the quiet number. Nobody puts it on the homepage, nobody brags about it in a launch tweet, and for a while you can ignore it because new signups paper over the gap. Then one quarter the signups slow down, and suddenly the rate at which people leave is the only thing that matters. This post walks through what churn rate actually measures, the formula behind it, why customer churn and revenue churn are not the same number, and why a figure that looks tiny on a monthly dashboard can quietly eat half your customer base in a year.

The formula is simpler than people expect

Customer churn rate is the share of customers who left during a period. That's it:

churn rate = customers lost in period / customers at start of period × 100

The two things that trip people up are both about the denominator. First, use the count you had at the start of the window, not the end. If you grew during the month, dividing by the larger ending headcount makes churn look better than it was, which is exactly the wrong way to flatter a number you're trying to control. Second, count whole customers, not logins or seats — this is "logo churn," one cancelled account equals one churned customer regardless of plan size.

Retention is just the mirror image inside the same base:

retention rate = 100 − churn rate

If 5% of your customers left, 95% stayed. They're not two different measurements; they're two ways of reading the same fraction. I keep both in front of me because they pull attention in opposite directions: churn is the leak, retention is the hull. You fix the leak, but you ship on the hull. You can run either reading instantly in the churn rate calculator by entering your starting count and the number you lost.

A worked example: 5% a month is not a rounding error

Say you start a month with 500 customers and 25 cancel. Plug it in:

25 / 500 × 100 = 5% monthly churn
retention = 100 − 5 = 95%

Five percent. It reads like a rounding error. Here's where it stops being one. Each month you keep 95% of who you had, so after twelve months you don't keep 95% — you keep 0.95 to the twelfth power, which is about 0.54. The honest annual churn figure is:

annual churn = 1 − (1 − 0.05)¹² ≈ 1 − 0.54 = 46%

Not 60% (you can't just multiply by 12, because the base shrinks every month as people leave), but close to half. At 5% monthly churn, you replace nearly half your customer base every year just to stand still. If you want to grow, your acquisition has to fill that hole first and then add on top. That's the part founders feel in their stomach the first time they see it compounded out.

There's a second number hiding in that same 5%. Average customer lifetime in months is the inverse of monthly churn:

average lifetime = 1 / 0.05 = 20 months

So your typical customer sticks around for 20 months. Cut churn to 4% and lifetime stretches to 25 months — a one-point change buys you five extra months per customer, a 25% lift in how long they pay you. That inverse relationship is why retention work has such outsized leverage near the low end: every point you shave off is worth more than the last.

Customer churn vs revenue churn

Counting accounts tells you how many relationships ended. It doesn't tell you how much money walked out the door. That's revenue churn, usually measured as MRR churn:

MRR churn = recurring revenue lost in period / recurring revenue at start × 100

The two diverge whenever the customers who leave aren't average-sized. Lose 25 of 500 customers, but they were all on your cheapest plan, and your 5% logo churn might be only 2% MRR churn — annoying, not dangerous. Flip it: lose three accounts out of 500 and one of them is a whale paying a quarter of your revenue, and logo churn looks calm at well under 1% while MRR churn is a five-alarm fire.

This is why "is churn up?" is a bad question on its own. Up which churn? A spike in logo churn driven by free-trial converts who never paid much is noise. A flat logo churn that hides one enterprise cancellation is a signal you'd miss entirely if you only watched the account count. Track both, side by side, and let the gap between them tell you whether you're losing customers or losing money.

Why retention is the number you actually manage

Churn is a result; retention is the thing you build. The framing matters because it changes what you do on Monday morning. "Reduce churn by 1 point" sounds like a defensive task — patch a leak. "Add five months to average customer lifetime" sounds like a product mandate — make the thing worth keeping. Same math, very different energy in a planning meeting.

Retention also compounds in your favor the way churn compounds against you. A cohort that stays 25 months instead of 20 doesn't just pay you 25% more; it has more time to expand its plan, refer others, and turn into a case study. The downstream economics of retention is why a small, durable improvement to monthly churn is often worth more than a flashy acquisition channel. If you're sizing that trade, the ROI calculator helps you put the retention experiment and the ad spend on the same scale.

Putting it to work

A practical loop: each month, compute logo churn and MRR churn from your start-of-period numbers. Annualize the monthly figure so you're never lulled by the small version of it. Read off the implied average lifetime. Then ask one question — did we lose customers or did we lose revenue? — and aim your next retention experiment at whichever answer is worse.

Don't quote the comforting monthly number to your board and the scary annual one to yourself. Pick the honest figure, which is almost always the compounded one, and let it anchor the conversation. A churn rate you can see clearly is a churn rate you can actually fight.

---

Made by Toolora · Updated 2026-06-13