Every subscription brand I look at has a churn number on a dashboard somewhere. Almost none of them have split that number into the two things it actually contains. There is the customer who decided they were done with you, and there is the customer whose card declined on a Tuesday while they were on holiday. Both show up in the same column. Only one of them is a marketing problem.
Failed payments drive 25 to 40% of total subscription churn across the industry. Stripe attributes roughly a quarter of all lapsed subscriptions purely to payment failure. That is a group of customers who still want your product, still have the habit, and still have a card that will probably authorise next week. Letting them lapse is the most expensive housekeeping error in DTC.
What Involuntary Churn Actually Costs You
Consumer subscriptions average 6.5% monthly churn, against 3.27% for subscription businesses overall. Strong physical product subscriptions sit around 4 to 5%. If a third of that number is involuntary, you are losing roughly 2% of your subscriber base every month to a billing mechanism rather than to a competitor, a price objection, or a product problem.
Run that against your own numbers. A brand with 4,000 active subscribers at 42 GBP per month is turning over 168k GBP in recurring revenue. Two per cent lost monthly to failed payments is 80 subscribers, 3,360 GBP of immediate revenue, and the entire remaining lifetime value of every one of those relationships. Replace them through paid acquisition at a 70 GBP CAC and you have spent 5,600 GBP buying back people who never left.
Voluntary churn is feedback. Involuntary churn is leakage. One tells you something about your product. The other tells you something about your plumbing, and plumbing is far cheaper to fix.
Why 13% of Your Recurring Charges Fail Every Month
Most operators assume declines mean the customer has run out of money or run out of interest. The data says otherwise. The majority of declines are soft, which means the same card on the same account will authorise on a later attempt. Here is what is actually happening when a recurring charge fails.
Insufficient funds
Around 44% of all payment declines. It is the single biggest cause and the most recoverable one. The customer gets paid on the 25th, your billing anchor sits on the 23rd, and the charge fails for two days every single month until the subscription dies. Timing beats persuasion here.
Expired cards
Roughly 10 to 12% of failures. Entirely predictable and entirely preventable. You know the expiry date on file. You can see it coming 60 days out. Most brands do nothing with that information until the charge has already failed.
Bank velocity and risk flags
Issuers flag recurring charges that look unusual against the customer's recent pattern. These come back as soft declines and frequently authorise on a second attempt, particularly when the retry is spaced rather than fired immediately three times in a row.
Stale billing details
The customer switched bank, got a replacement card after fraud, or moved house and the billing postcode no longer matches. The relationship is intact. The record is not. These need a human-facing prompt, not another retry.
The Retry Schedule Your Subscription App Ships With Is Not Good Enough
Default retry logic in most Shopify subscription apps recovers 40 to 50% of failed payments. A properly built dunning system recovers 70 to 85%. That gap is not a tooling gap. Nearly every app on the market can be configured to do this. The gap exists because nobody ever opened the settings.
Four attempts across 14 days is the cadence that works for most DTC subscription brands. Day 0 catches transient network and gateway errors. Day 3 catches reset daily limits. Day 7 reaches subscribers paid weekly. Day 14 lands near fortnightly payroll. Network-informed retry timing, where the schedule adapts to the specific decline code and the issuer, outperforms fixed intervals by 10 to 20 percentage points.
The other number that should change how you build this: around 90% of all recovered transactions happen within the first 10 days of the original failure. Effort spent on day 21 is almost entirely wasted. Effort spent on days 0 to 10 is where the revenue is. Front-load the retries, front-load the messaging, and stop running 30 day dunning cycles that annoy people and recover nothing.
The Dunning Sequence That Recovers Revenue
A retry is a machine talking to a bank. A dunning sequence is you talking to a customer. You need both, and the messaging is where most brands do real damage by sending four identical red-bordered emails that read like a debt collector. Here is the structure that works.
Pre-dunning, 7 days before the charge
Fires only on cards expiring within the window or on accounts with a prior failure. Friendly, low urgency, one job: update the card before anything breaks. This is the cheapest recovery you will ever run because the subscription never lapses in the first place.
Day 0, the charge fails
Short, calm, no alarm language. Tell them the payment did not go through, tell them you will try again automatically, and give a one-tap link to update the card with no login required. Forcing a password reset at this moment kills more subscriptions than the decline did.
Day 3 and day 7, the reminders
Shift the framing from billing to the product. Name what they are about to miss: the refill that was due, the member price, the queue position. Add SMS from day 7 onwards. Email engagement in dunning sequences drops to 20 to 30% by that point while SMS open rates sit above 90%.
Day 13 to 14, the last attempt
Send this before any suspension, not after. Make the consequence explicit and make the alternative explicit too. Offer pause as a genuine option rather than only offering payment. A paused subscriber is a future customer. A cancelled one is a new acquisition cost.
The Prevention Layer Nobody Switches On
Recovery is the expensive half. Prevention is the half that runs silently and costs you nothing once it is configured. Start with a card account updater, which syncs replacement card numbers and new expiry dates directly from the networks. Recurly B2C data shows authorisation rates improve by roughly 2 percentage points once it is enabled. On a few thousand active subscribers that is real money every month, collected without the customer ever seeing a form.
Then look at your billing anchor. If a meaningful share of your base is declining for insufficient funds, you are charging on the wrong day. Letting subscribers choose their billing date, or defaulting new subscriptions to land shortly after common UK pay dates, removes a whole category of failure before it happens.
Finally, treat pause as infrastructure rather than as a concession. Around three out of four subscribers who pause eventually come back, and pause usage grew 337% year on year. A cancellation flow that offers a real alternative saves 15 to 25% of attempts. Both of those numbers beat anything your acquisition channel is doing right now.
What This Looks Like in Practice
A wellness brand I work with was reporting 7.1% monthly subscription churn and had briefed their team on a retention campaign to fix it. Before anyone wrote a word of copy, we split the churn number. Thirty-eight per cent of it was involuntary. They were running the default three retries across five days with a single notification email, and recovering 44% of failed charges.
We moved the retries to a day 0, 3, 7, 14 schedule, enabled the account updater, added a pre-dunning email for cards expiring inside 60 days, rewrote the sequence to lead with the product rather than the invoice, and added SMS from day 7. Recovery went from 44% to 79% over the following two billing cycles. Reported churn fell to 5.2% with no change to the product, the price, or the acquisition mix.
The retention campaign never ran. It was not needed. The customers had never decided to leave.
Inside the system
How we build this for brands
When we take a subscription brand on, the first thing we build is a reporting layer that splits churn into voluntary and involuntary automatically, pulling live from Shopify and the subscription app so the number is in front of the operator weekly rather than being reconstructed in a spreadsheet once a quarter. Sitting on top of that, we deploy the dunning and pre-dunning sequences in Klaviyo with AI, segmented by decline reason rather than sending everyone the same four emails, and wire SMS into the back half of the window where email engagement collapses.
Alongside it runs a reporting agent that watches recovery rate, authorisation rate, and billing anchor distribution, and flags the week they move rather than the month after. Part of this runs live for portfolio brands today; the full system is what we deploy when we take a brand on.
Subscription Audit
Find Out How Much Of Your Churn Is Actually Failed Payments
I will split your churn into voluntary and involuntary, pull your real recovery rate, and show you exactly where your billing setup is cancelling customers who never wanted to leave. Numbers first, then the fix list.
Book Your Subscription AuditFrequently asked questions
What is involuntary churn in a subscription business?
Involuntary churn is when a subscriber is lost because their payment failed, not because they chose to leave. The card expired, the bank declined it, funds were short, or the billing details went stale. Industry data puts involuntary churn at 25 to 40% of total subscription churn, and Stripe attributes roughly 25% of lapsed subscriptions purely to payment failure. It is the only form of churn where the customer still wants your product.
What percentage of recurring payments fail each month?
Recurly data shows an average of 13% of recurring transactions decline every month. The largest single cause is insufficient funds at roughly 44% of all declines, followed by expired cards at 10 to 12%. Most of those declines are soft, meaning the same card will authorise on a later attempt if your retry logic is set up properly.
What is a good failed payment recovery rate for DTC subscriptions?
Default retry logic in most Shopify subscription apps recovers 40 to 50% of failed payments. An optimised dunning system that combines smart retry timing, a card account updater, and a proper notification sequence recovers 70 to 85%. If you do not know your current recovery rate, that number is the first thing to pull before changing anything else.
How many times should you retry a failed subscription payment?
Four attempts spread across 14 days works for most DTC subscription brands: an immediate retry on day 0 to catch network errors, day 3 to catch reset daily limits, day 7 to reach weekly paid subscribers, and day 14 to align with fortnightly payroll. Around 90% of all recovered transactions happen within the first 10 days, so front-load the effort rather than retrying endlessly at the back end.
Does a card account updater actually make a difference?
Yes, though the gain is quieter than most vendors imply. Recurly B2C data shows authorisation rates improve by roughly 2 percentage points once an account updater is enabled. On a subscription base of several thousand active members that is meaningful recurring revenue every month, and it works silently without the customer ever touching a form.
Should you offer a pause instead of letting a failed payment cancel?
Pause is one of the strongest retention tools available. Around three out of four subscribers who pause eventually return, and Recurly recorded a 337% year on year increase in pause usage. If a payment keeps failing because of genuine cash pressure, a pause keeps the relationship alive where a cancellation ends it. Cancellation flows with a real alternative typically save 15 to 25% of attempts.
About the author
Caner Veli is a DTC operator who has helped 350+ brands fix broken growth engines. He built Liquiproof from zero to 3,000+ global retailers in under 6 years. He now runs the same playbook, supported by AI systems he built himself, for DTC and CPG brands.