Subscription Churn Benchmarks by Industry

Every subscription operator eventually asks the same question: is my churn normal? The honest answer is that "normal" depends on what you sell, who you sell it to, and what you charge — and most benchmark numbers floating around blog posts cannot survive a source check. This page sticks to figures that can. Every number below comes from the published data assembled on our failed payment recovery benchmarks page: research published by Recurly, and Churnkey's analysis of Stripe data. Where a figure came from, it says so. Where we do not have a published figure, this page does not invent one.

Use these churn benchmarks by industry as a range to locate yourself in, not as a grade. A benchmark tells you whether your problem is typical, better than typical, or worse than typical. It does not tell you what to fix — that comes from your own numbers, and the last section covers how to use benchmarks without lying to yourself.

What these benchmarks actually measure

Two different measurements run through this page, and confusing them is the most common way benchmarks get misused.

Annual churn is the share of subscribers a business loses over a year, for all reasons combined — deliberate cancels and payment failures together. Involuntary churn is the slice of that loss caused by failed payments rather than decisions: the subscriber did not choose to leave, their card failed, the retries failed, and the subscription ended anyway. The industry table below shows both, plus involuntary churn as a share of all churn — the number that tells you how much of your loss is a billing problem instead of a product problem.

The distinction matters because the fixes are different. Voluntary churn is answered with product, pricing, and cancel flows. Involuntary churn is answered with retry timing, card updates, and outreach. Read only the combined annual figure and you cannot tell which problem you have. If you need the formulas behind your own numbers first, start with how to calculate churn rate — the comparison only means anything if your churn rate was calculated the same way every month.

Churn benchmarks by industry

Churnkey partnered with Stripe to analyze subscription transaction data — Stripe processed $1.4 trillion across 200 million subscriptions in 2024 — and published the split of voluntary versus involuntary churn by industry (Churnkey, Involuntary Churn Benchmarks; underlying annual churn rates from Stripe). The figures below mirror that published table exactly:

Industry	Annual churn	Annual involuntary churn	Involuntary as % of all churn
Digital goods	38%	11%	29%
Business services	40%	10%	26%
Personal services	36%	9%	25%
Education	40%	10%	24%
Merchandise	39%	9%	23%
SaaS	38%	8%	22%
Travel and lodging	43%	8%	18%
Leisure	36%	6%	17%
Insurance	37%	3%	9%

Three readings worth taking from the table:

• Total churn clusters tightly; the involuntary share does not. Annual churn sits in a narrow band across all nine industries, from the mid-thirties to the low forties. The involuntary share runs from 29% for digital goods down to 9% for insurance. Two businesses with the same headline churn rate can have completely different problems — one losing subscribers who decided to leave, the other losing subscribers whose payments failed.
• Digital goods carry the highest involuntary share at 29% (Churnkey analysis of Stripe data). Low-friction signups, small recurring charges, and prepaid or lightly used cards make payment failure a large part of the loss. If you sell digital goods, billing recovery is not a back-office detail; it is one of your largest retention levers.
• Insurance is the outlier at the other end, at 9% (Churnkey analysis of Stripe data). Longer contracts, larger payments, and payment methods customers actively maintain all push the involuntary share down. Do not benchmark a consumer subscription against insurance and conclude your billing is broken — the payment behavior underneath the two industries is different by design.

B2C versus B2B: 24% versus 16%

The same Churnkey analysis of Stripe data splits the involuntary share by customer type: in B2C businesses, 24% of all churn is involuntary — lost to payment failures — while in B2B businesses the figure is 16% (Churnkey analysis of Stripe data).

The gap makes mechanical sense. Consumer subscriptions are paid on personal cards that expire, get replaced, and hit limits. Business subscriptions are more often paid on company cards or arrangements someone in finance actively maintains, and a failed business payment is more likely to be noticed and fixed by the customer. The takeaway: a B2C operator should expect roughly a quarter of churn to be a payments problem, while a B2B operator with the same involuntary share as a B2C peer is underperforming their segment.

Price point changes the mix

The lower the price, the larger the share of churn that is involuntary. At an average order value under $10, about 35% of all churn is payment failure (Churnkey analysis of Stripe data).

The mechanics are not mysterious. A small charge gets the least attention from the cardholder: nobody reorganizes their finances around a charge smaller than lunch, prepaid and low-balance cards are more common at low price points, and a subscriber paying very little is the least likely to act on a dunning email for a product they barely think about. The consequence: at low price points involuntary churn is not a rounding error — at roughly a third of all churn, it is a main component, and recovery work aimed at it is retention work, not collections work.

The size of the problem

Two Recurly figures frame the overall scale. Recurly estimates involuntary churn was a $129B problem for the subscription industry in 2025 (Recurly). And Recurly publishes the working range most operators plan against: 20–40% of all subscriber churn is involuntary (Recurly).

How to use benchmarks without lying to yourself

Benchmarks fail in predictable ways. These are the failure modes worth guarding against:

• Do not compare across the wrong row. Compare your involuntary share to your industry's, your segment's (B2C or B2B), and your price point's — not to the most flattering number on the page. An operator who benchmarks a low-priced B2C product against the insurance row is not benchmarking; they are shopping for comfort.
• Do not mix measurement types. Annual churn, monthly churn, voluntary churn, and involuntary share are four different numbers. Most benchmark misuse is a units error: your monthly figure held up against someone else's annual figure, or your total churn against someone else's involuntary-only figure.
• Do not treat the benchmark as the target. Being at your industry average means your billing leaks at the average rate. The involuntary slice is recoverable loss by definition — subscribers who wanted to stay. "Typical" is a diagnosis, not a goal.
• Do not quote a benchmark you cannot source. Every figure on this page names its publisher. If a number in your board deck cannot do the same, it is decoration. Replace it with your own measured figure or drop it.
• Do not let a good benchmark stop the investigation. An involuntary share inside the published 20–40% range (Recurly) still means a fifth to two-fifths of your losses came from failed payments. The benchmark tells you the problem is normal. It does not tell you the problem is small.

From benchmarks to your own numbers

The useful sequence is: locate yourself in the table, then replace the benchmark with your own measurement as fast as you can. Compute your involuntary share directly — of the subscribers you lost last quarter, how many ended over a failed payment rather than a cancel click — and your recovery rate on the payments that failed. Then put your own volume through the recovery calculator: you set the recovery-rate assumption, the formula is printed on the page, and nothing is claimed for you.

That is also the honest way to evaluate recovery software, including ours. Swipe Again is failed-payment recovery software for subscription businesses — decline-code-aware smart retries (up to 6 attempts over about 16 days, hard declines never retried), branded dunning emails and card-update pages, SMS in the sequence, and cancel deflection — for $179/mo flat, no percentage of recovered revenue, on a 14-day trial with the card upfront and a 90-day backfill on connect. Its guarantee, in exact form: "Recover $179 in 30 days, guaranteed — or Swipe Again is free the next 30 days." We have no first-party recovery data yet, which is why this page is built from Recurly and Churnkey figures instead of ours.

The bottom line

Annual churn clusters in a narrow band across industries, but the involuntary share runs from 29% down to 9% (Churnkey analysis of Stripe data). B2C businesses lose 24% of their churn to payment failures against 16% for B2B, low price points push the involuntary share to about 35% (Churnkey analysis of Stripe data), and Recurly puts the industry-wide cost at an estimated $129B for 2025, with 20–40% of all churn involuntary (Recurly). Find your row, your segment, and your price point — then go measure your own involuntary share, because that is the number you can actually move.

Frequently asked questions

What is a normal churn rate for a subscription business?

In the Churnkey analysis of Stripe data, annual churn by industry runs from 36% for personal services and leisure to 43% for travel and lodging, with most industries clustered between 37% and 40%. Normal depends heavily on industry, customer type, and price point, so compare against your own row of the table rather than a single headline number.

How much of churn is involuntary?

Recurly publishes a working range of 20–40% of all subscriber churn being involuntary. The Churnkey analysis of Stripe data puts the involuntary share at 24% for B2C businesses and 16% for B2B businesses, ranging by industry from 29% for digital goods down to 9% for insurance.

Why do low-priced subscriptions lose more subscribers to failed payments?

According to the Churnkey analysis of Stripe data, at an average order value under $10, about 35% of all churn is payment failure. Small charges get the least attention from cardholders, are more often placed on prepaid or low-balance cards, and subscribers paying very little are less likely to act on a payment problem for a product they rarely think about.

Where do these churn benchmarks come from?

The industry table, the B2C versus B2B split, and the price-point figures come from Churnkey's published analysis of Stripe data. The estimated $129B cost of involuntary churn for 2025 and the 20–40% involuntary range come from Recurly research. Swipe Again has no first-party benchmark data yet and publishes none on this page.

What should I do if my involuntary churn is above my industry benchmark?

Treat the gap as recoverable loss, not as a verdict. Check your failure mix first: timing-driven failures respond to decline-aware retry scheduling, while expired and replaced cards respond to card-update outreach. Then measure your recovery rate on failed payments and compare your before and after, rather than comparing yourself to a benchmark again.