Involuntary churn and dunning: how failed payments quietly cost SaaS about 9 percent of revenue

Involuntary churn is revenue you lose when a subscription cancels because a payment failed, not because the customer chose to leave. The usual causes are an expired card, insufficient funds, or a bank decline. It accounts for 20 to 40 percent of total SaaS churn and quietly costs the average subscription business about 9 percent of recurring revenue a year.

Most founders never see it. The loss gets booked as product or voluntary churn, so the leak hides in the dashboard and nobody owns it.

The fix is dunning: an automated retry schedule plus payment-update messaging that recovers most failed charges before the customer notices anything went wrong.

What is involuntary churn

Involuntary churn is the loss of a paying subscriber because a recurring payment failed, not because they decided to cancel. They never intended to leave. The card expired, the bank declined the charge, or the account was short that day.

That distinction is the whole game. Voluntary churn is a product or pricing problem you fix with retention work. Involuntary churn is a billing glitch you fix with timing and tone.

Here is the split, side by side.

	Voluntary churn	Involuntary churn
Cause	Customer chooses to cancel	Payment failed
Customer intent	Wants to leave	Wants to stay
Recoverable	Rarely	Mostly (about 70 percent)
Fix	Product, pricing, value	Retries, pre-dunning, card updates

You measure both off the same base, but you fix them with completely different tools. Lumping them into one churn rate is the first mistake. It hides the cheapest revenue you will ever recover.

The formula is simple:

Involuntary churn rate = (customers lost to failed payments in period / total customers at start of period) x 100

And the share that matters most:

Involuntary share of churn = (involuntary churned customers / total churned customers) x 100

The 9 percent leak nobody books

The average subscription business loses about 9 percent of recurring revenue a year to failed payments. Companies running no dunning automation lose 9 to 12 percent. That single number erases a full month of growth.

It stays invisible because of how the loss gets recorded. When a card fails and the subscription cancels, most billing exports just show a churned customer. There is no flag that says “this one wanted to stay.” So the 9 percent gets folded into total churn, blamed on the product, and assigned to no one.

You cannot fix a number you do not measure. The highest-ROI move here is not buying a dunning tool. It is making involuntary churn and recovery rate first-class metrics so the leak surfaces before any retry cadence runs.

Mowt splits involuntary from voluntary churn analytics straight from your Stripe data, so you see how much MRR is leaking to failed payments before deciding what to fix. Once the number is visible, the math gets uncomfortable fast.

What a 10M ARR company actually loses

Take a company at $10,000,000 ARR with a 12 percent failed-payment rate and a single retry attempt, which recovers about 25 percent of failures.

Annual revenue lost = ARR x failed-payment rate x (1 minus recovery rate)

Run it:

$10,000,000 x 0.12 x (1 minus 0.25) = $10,000,000 x 0.12 x 0.75 = $900,000

That is $900,000 a year walking out the door, climbing toward $1,200,000 at the high end of the failed-payment range. Now lift recovery from 25 percent to 70 percent, which is what a full dunning stack delivers:

$10,000,000 x 0.12 x (1 minus 0.70) = $360,000 lost.

Same failure rate. Same product. The difference is $540,000 a year, recovered by retrying smarter and prompting card updates. No new customers, no price changes. This is the cleanest revenue most teams ignore, and it shows up directly in your net revenue retention.

What percentage of churn is involuntary

For subscription businesses, involuntary churn runs 20 to 40 percent of total churn. For SaaS specifically, Stripe data puts it around 22 percent. The exact share depends on your model.

Segment	Involuntary share of churn
SaaS overall (Stripe)	approx 22%
B2C subscriptions	approx 24%
B2B SaaS	approx 16%
High-decline segments	40 to 48%
Subscription consensus range	20 to 40%

If a third of your churn is involuntary and you are not measuring it, a third of your churn problem is fixable with billing logic, not product work. That reframes the roadmap. Before you build the next retention feature, check what share of your lost logos simply had a card decline.

Why payments fail

Most failures are not fraud or fake cards. They are mundane, and that is good news, because mundane is recoverable.

Insufficient funds: the single largest cause at about 41 percent of failed subscriptions (Churnkey, across 5 million failed subs). The money is there a few days later.
Expired or replaced cards: the most recoverable failure type, at 80 to 90 percent recovery potential. A reissued card just needs an update.
Bank and fraud-system declines: a smaller slice, often clearing on a retry once the bank’s risk window passes.

The 41 percent insufficient-funds figure is why retry timing matters more than retry count. People get paid, pending transfers clear, and credit limits reset on a rhythm. Hit the card on the right day and it clears on its own.

What dunning is, and why it is customer success

Dunning is the automated process of retrying failed charges and prompting customers to update their payment details. The word sounds like collections. Run it like collections and you will lose customers who wanted to stay.

Good dunning is a customer-success function. The goal is to quietly fix a billing problem before a happy customer is cut off. Tone and timing do more work than engineering here.

The relationship is simple. Involuntary churn is the problem. Dunning is the solution.

The retry schedule that actually works

A staggered cadence beats daily retries. Retry on Day 1, Day 3, Day 5, and Day 7. That spread straddles paydays, lets pending transfers clear, and gives credit limits time to reset, which is exactly why insufficient-funds declines clear on their own.

This recovers about 58 percent of failed payments through retries alone, before you contact the customer at all (Recurly 2025). Each attempt has sharply diminishing returns, so front-load them.

Retry attempt	Approx success rate
1st attempt	approx 35%
2nd attempt	approx 15%
3rd attempt	approx 8%

The gap between smart and naive retries is enormous. Smart retry logic recovers about 68 percent of failed payments. A single retry attempt recovers about 23 percent (Churnkey). Same failures, triple the recovery, purely from cadence.

One rule: match timing to the decline reason. Soft declines like insufficient funds warrant quick retries. Hard declines like an expired card need a card update first, because retrying a dead card a hundred times recovers nothing.

Pre-dunning: recover before the charge fails

Pre-dunning prompts customers to update an expiring or about-to-fail card before the payment fails. It recovers an extra 15 to 22 percent of at-risk revenue (Stripe billing data).

This is the cheapest recovery of all, because the customer is still active and happy when you reach them. You are not chasing a failed charge. You are sending a friendly “your card on file expires next month” note.

It maps to the most recoverable failure type. Expired cards have 80 to 90 percent recovery potential, and pre-dunning prevents most of those failures from ever happening. Pair it with an automatic card-account updater service and a large chunk of your expiry-driven churn disappears.

The full recovery stack

Layering retries, pre-dunning, and a short email or SMS sequence recovers about 70 percent of all detected involuntary churn (Churnkey). No single layer carries it. They stack.

Retries (Day 1, 3, 5, 7): about 58 percent, automatic, no contact.
Pre-dunning (update card before failure): an extra 15 to 22 percent.
Email and SMS over about 2 to 3 weeks: closes the tail.
Automatic card updaters: kill expiry-driven failures at the source.

Do not over-index on email. Dunning email recovery decays fast: email 1 recovers about 2.8 percent, falling to about 1.5 percent by email 5 (Churnkey). Emails complement retries. They do not carry recovery.

Here is the effect on your books. A 9 percent gross involuntary leak at 70 percent recovery nets to roughly 2.7 percent:

Effective churn after recovery = gross involuntary churn x (1 minus recovery rate) = 9% x (1 minus 0.70) = 2.7%

How your recovery rate compares

The median SaaS failed-payment recovery rate is just 47.6 percent (Slicker 2025). Most companies recover less than half of failed charges. That median is the opportunity: matching best-in-class retry timing is a near-instant revenue win.

Segment	Failed-payment recovery rate
SaaS median (Slicker 2025)	47.6%
Enterprise SaaS (ACV above $10K)	52 to 58%
SMB SaaS (ACV below $1K)	38 to 47%
Full stack: retries + pre-dunning + email	approx 70%

Enterprise recovers more because larger accounts have steadier payment methods and a human to call. SMB recovers less, so SMB-heavy businesses have the most to gain from automation. If you are below 47.6 percent, the recovery requires no product work.

The ROI is hard to argue with. Effective dunning programs report roughly 10x to 15x return, with many seeing payback inside the first month (Baremetrics aggregate: $1,350,000 recovered across 148 customers, 82 percent paying back within the month).

How to reduce involuntary churn

Reducing involuntary churn is a five-step playbook, and the first step is measurement, not tooling.

Measure it separately. Split involuntary from voluntary churn so the 9 percent leak is visible. Track involuntary churn rate and recovery rate as real-time metrics. You cannot manage a number that hides inside total churn.
Fix retry timing first. Move to a Day 1, 3, 5, 7 cadence. This recovers about 58 percent automatically and costs nothing but configuration.
Add pre-dunning. Prompt customers to update expiring cards before they fail, for another 15 to 22 percent.
Run a short dunning sequence. Two to three weeks of well-toned email and SMS to close the tail. Sound like support, not a debt collector.
Turn on card updaters. Automatic account updater services kill expiry-driven failures before they reach your retry logic.

Start with step one. The retry cadence and pre-dunning only pay off once you can see what they are recovering, which is why surfacing involuntary churn in real time, alongside your cohort retention and LTV, makes everything downstream measurable.

The leak is fixable. It is also invisible by default. Make it a metric, match best-in-class retry timing, and you recover most of a month of growth that was quietly walking out the door.

FAQ

What is involuntary churn?

Involuntary churn is when a paying subscriber is lost because a recurring payment failed, not because they decided to cancel. The usual causes are expired or replaced cards, insufficient funds, and bank or fraud-system declines. Because the customer never intended to leave, most of it is recoverable through dunning.

What percentage of churn is involuntary?

For subscription businesses, involuntary churn is typically 20 to 40 percent of total churn. For SaaS specifically, Stripe data puts it around 22 percent, with B2C closer to 24 percent and B2B closer to 16 percent. In high-decline segments it can reach 40 to 48 percent.

How much revenue do failed payments cost SaaS companies?

The average subscription business loses about 9 percent of recurring revenue a year to failed payments, and companies with no dunning automation lose 9 to 12 percent. For a $10,000,000 ARR company that is roughly $900,000 to $1,200,000 a year, enough to erase a full month of growth.

What is the best dunning retry schedule?

A staggered cadence on Day 1, Day 3, Day 5, and Day 7 beats daily retries, because it straddles paydays, lets pending transfers clear, and resets credit limits, the main reasons insufficient-funds declines (about 41 percent of failures) clear on their own. This recovers roughly 58 percent before any customer contact. Match timing to the decline reason: soft declines warrant quick retries, while a hard decline like an expired card needs a card update first.

How do you reduce involuntary churn?

Layer four tactics: an optimized retry cadence (Day 1, 3, 5, 7) for about 58 percent automatic recovery, pre-dunning that prompts a card update before failure for another 15 to 22 percent, a short dunning email and SMS sequence, and automatic card-account updater services. Together these recover about 70 percent of detected involuntary churn. Track involuntary churn and recovery rate as real-time metrics so the leak stays visible.