Subscription Churn Rate

Subscription churn rate: how to calculate it, and what it’s really telling you

Churn feels small month to month and is enormous in aggregate. Lose 8% of subscribers a month and simple compounding means roughly 63% of the subscribers you started the year with are gone by its end — before counting a single new signup you also have to churn-proof. That arithmetic is why churn rate, not signup rate, decides whether a subscription program compounds or just treads water expensively.

The formula

Customer churn rate for a period is: subscribers lost during the period ÷ subscribers at the start of the period. Thirty cancellations from a base of 500 in June is a 6% monthly churn rate. Two refinements make the number honest. Exclude the customers who joined during the period from the denominator (or run the math by cohort), so a big acquisition month doesn’t flatter the rate. And decide explicitly how you count pauses and skips — a paused subscriber isn’t churned, but a “pause” that never returns is; track them as their own state rather than letting them hide in either bucket.

Customer churn vs revenue churn

Counting people and counting money answer different questions. Revenue churn measures the recurring revenue lost — cancellations plus downgrades, skipped shipments and reduced quantities. A program can hold customer churn steady while revenue churn quietly climbs, because subscribers are downshifting without leaving. Track both: customer churn tells you whether the relationships survive; revenue churn tells you whether the P&L does.

What a “good” churn rate is — the honest answer

Any single benchmark number you find online deserves skepticism: churn varies enormously with category, price point, shipment cadence and how easy the brand makes pausing. A supplement on a monthly cycle, a coffee bag every two weeks and a quarterly apparel box live in different churn worlds. The comparisons that actually mean something are: your own trend (is this quarter’s churn below last quarter’s), your cohorts (do customers who joined in May survive better than those from January), and your voluntary/involuntary split — because those demand completely different fixes. If you want an external yardstick, benchmark against your category and cadence, not against “subscription businesses” as a whole.

Read the curve, not the average

Plot survival by subscription month and nearly every DTC program shows the same shape: a steep cliff across the first two or three shipments, then a plateau of long-term keepers. That shape is the diagnosis. A blended monthly rate of 6% might mean healthy long-term retention with a brutal early cliff — an onboarding problem, the kind we unpacked in reducing early churn — or steady erosion across all tenures, which points at cadence and product fit. Same average, different disease, different cure.

Split voluntary from involuntary before fixing anything

Some cancellations are decisions. Many aren’t — the card expired, the payment failed, the retry gave up, and a subscriber who wanted to stay was dropped by billing mechanics. That involuntary share is the first thing to measure, because it’s the cheapest churn to fix: dunning management — pre-expiry warnings, smarter retry timing, one-tap card updates — recovers revenue from people who never chose to leave. The voluntary remainder is a lifecycle problem with its own causes; we cataloged them in why subscribers cancel, and the fixes run from onboarding through cadence control to a save-first cancellation flow.

The fixes, in payback order

First, recover failed payments — pure leak, no persuasion required. Second, fix the first-90-days experience, because the cliff is where the volume is: teach the product, deliver early value, get shipment two right. Third, add cadence control — pause, skip, swap — so “too much product” becomes an adjustment instead of an exit. Fourth, build a cancellation flow that offers the right alternative before processing the cancel. Each of these bends the curve at a different point, which is why a subscription retention program sequences them rather than picking one.

Why a point of churn is worth so much

Churn compounds against you, so its reduction compounds for you. Every point of monthly churn you remove extends the average subscriber lifetime — and because those retained months arrive with no acquisition cost attached, they flow almost directly into customer lifetime value. Run the arithmetic on your own base: the revenue difference between your current churn rate and one point lower, carried across a year, is usually the strongest business case in your backlog.

Frequently asked questions

How do you calculate subscription churn rate?

Divide subscribers lost in a period by subscribers at the start of that period (30 lost from 500 = 6% monthly churn). Track revenue churn alongside it, count pauses as their own state, and run the number by cohort for an honest view.

What is a good subscription churn rate?

There’s no universal number — churn varies by category, price and shipment cadence. Judge yourself on trend (falling is good), cohort survival, and how much of your churn is involuntary, because failed-payment churn is the share you can fix fastest.

We diagnose and fix churn — dunning, onboarding and save flows — through our subscription retention service. Request an Audit →