How to use Dunning & Payment Recovery
Failed payments quietly cost subscription brands 20-40% of monthly churn. This guide walks through the retry logic, pre-dunning comms, and card-updater services that recover most of it.
Dunning & Payment Recovery
The retry logic and customer communication playbook that recovers failed subscription payments before they turn into churn.
Dunning is the structured process of retrying a declined card, notifying the customer, and updating payment credentials so a subscription doesn't lapse over a fixable billing problem. It sits at the intersection of payments engineering and lifecycle marketing: smart retry schedules handle the bank-side recovery, while pre- and post-failure emails handle the customer-side recovery.
For a DTC subscription brand, dunning is usually the single highest-ROI churn lever. A typical Shopify subscription store sees 5-9% of monthly charges fail on the first attempt; without a recovery flow, every one of those becomes involuntary churn. A well-tuned playbook recovers 55-75% of them.
Most subscription brands underinvest here because the symptom hides inside the billing dashboard rather than in the CRM. A cancellation shows up as a clear event with a reason code; a failed renewal just quietly stops generating revenue. Until you instrument it, you don't feel it.
The other reason it's neglected is that dunning lives in three teams at once — payments, retention, and lifecycle email — and none of them owns the full loop. This guide treats it as a single system: the retry schedule, the comms sequence, the card-updater layer, and the metrics that prove it's working.
Designing the retry schedule
A retry schedule is the sequence of automatic re-charge attempts after a payment fails. The naive version is "retry every day for a week". The smart version routes attempts around the failure reason code and the customer's payday pattern.
Decline reasons fall into two buckets. Hard declines (lost card, stolen card, invalid account) will never succeed on retry — you need a new card, not another attempt. Soft declines (insufficient funds, issuer unavailable, velocity limit) recover on their own within days as balances refresh or bank-side throttles reset.
A workable default for soft declines: retry on day 1, day 3, day 5, and day 7, with one final attempt around day 14 to catch the next payday cycle. Hard declines should skip retries entirely and trigger the comms flow immediately, because the only path forward is the customer updating their card.
Don't retry on the day of failure
Re-attempting a card within an hour of decline almost never works — the issuer's velocity controls are still active, and repeated declines in 24 hours can flag the merchant for elevated processor scrutiny. Wait at least 24 hours before the first retry.
The pre-dunning and dunning comms sequence
Pre-dunning is the email you send before a payment is even attempted, when you can see a card is about to expire or has been flagged by your processor's account-updater service. It's the highest-leverage email in the entire flow because the failure hasn't happened yet — the customer is still in a positive frame, not a recovery one.
Post-failure comms should run alongside the retry schedule, not after it. A typical sequence: email 1 on the day of first failure ("we couldn't process your payment, here's a one-click update link"), email 2 on day 4, SMS on day 7 if you have consent, and a final "your subscription will pause" notice on day 14.
Recovery rate by dunning email touchpoint
The pattern almost every brand sees: the first email recovers the most absolute volume, but pre-dunning has the highest per-send conversion because you're catching a customer who hasn't failed yet. SMS pulls disproportionate weight on day 7 when the email reminders have already been ignored.
Card updater services and network-level recovery
Visa and Mastercard both run account-updater networks (VAU and ABU) that let merchants automatically refresh stored card credentials when a customer's bank reissues a card — new expiry date, new card number after a fraud replacement, or a reactivated account. Stripe, Adyen, and Recharge all expose this as a feature, but it's often off by default.
Turning on network tokenisation alongside the account updater is the next lever. Network tokens replace the raw PAN with a token issued by the card network, which gets refreshed automatically and typically lifts authorisation rates by 2-4 percentage points across the whole subscription book — not just on the failure path.
Typical recovery rates by dunning layer (DTC subscription, soft declines)
| Recovery layer | Share of failures recovered | Setup effort | Notes |
|---|---|---|---|
| Smart retry schedule only | 30-40% | Low | Default in most billing platforms |
| + Pre-dunning email (card expiring) | +8-12 pp | Low | Requires processor expiry feed |
| + Post-failure email sequence | +12-18 pp | Medium | Lifecycle tool integration |
| + SMS touchpoint | +3-6 pp | Medium | Consent-gated; high per-send cost |
| + Account updater (VAU/ABU) | +5-9 pp | Low | Often a processor toggle |
| + Network tokenisation | +2-4 pp | Medium | Lifts baseline auth rates too |
| Full stack combined | 65-78% | — | Realistic ceiling for soft declines |
The compounding matters: each layer recovers a slice of what the previous layer missed, not the original failure pool. A brand running only smart retries leaves 60-70% of failures on the table — most of which a $0-marginal-cost processor toggle would catch.
Measuring dunning performance
The headline metric is recovery rate: of all failed charges in a cohort month, what share were eventually collected within 30 days. Track it by failure reason, by retry attempt number, and by comms touchpoint. The reason-code split is what tells you whether you have a retry problem or a customer-action problem.
Tie dunning into the broader Churn Reduction Playbook by reporting involuntary churn separately from voluntary churn. The distinction matters because the levers are completely different — see Voluntary vs Involuntary Churn for the upstream framing — and blending them hides the fact that a billing improvement is usually faster than a retention improvement.
The 90-day quick win
If you've never tuned dunning: turn on your processor's account updater this week, add a pre-dunning email for cards expiring in 30 days, and split your retry schedule by hard vs soft decline. Most brands see involuntary churn drop by a third inside one quarter.
Frequently asked questions
Dunning historically referred specifically to the customer-facing communication after a failed payment — the reminder emails and notices. Payment recovery is the broader system that includes the retry logic, card updaters, and network-level tools. In modern usage the terms are used interchangeably.
More than five attempts in a single failure cycle starts hurting you. Card networks track repeated declines, and excessive retries on the same card can elevate your merchant risk profile with the acquiring bank. Four to five intelligently-spaced attempts is the sweet spot.
No. A hard decline (lost card, stolen card, closed account, do-not-honor) will not succeed on a re-attempt — the issuer has flagged the credential. Skip retries entirely and route directly to the comms flow asking the customer to update their card.
A well-tuned stack — smart retries plus pre-dunning, post-failure emails, account updater, and network tokens — recovers 65-78% of soft-decline failures within 30 days. Hard-decline recovery depends entirely on whether customers update their card, which typically lands at 25-40%.
Failed-payment churn is the textbook example of involuntary churn — the customer didn't choose to leave, a billing system event ended their subscription. That's why it's the highest-ROI churn lever: the customer still wants the product, you just need to collect the money.
Yes, but only as a later-stage touchpoint after email reminders have been ignored. SMS on day 7 typically recovers 10-15% of remaining failures at a per-message cost that's easily covered by a single restored subscription. Skip SMS on day 1 — email handles that volume more cheaply.
If you're on Recharge, Bold, or Skio, it's usually a single toggle in the payment settings — they pass the request through to Stripe or your processor. For native Shopify Subscriptions, account updater runs automatically through Shopify Payments where supported by the card network and issuer.
Pre-dunning is an email sent before a payment is attempted, when you can see a card will fail — typically because it's expiring within 30 days. Your processor exposes the expiry date on the saved card object; query stored cards weekly and trigger an update-your-card email for any within the threshold.
Two weeks is the typical window. By day 14 you've exhausted soft-decline retries, run the full email sequence, and given the customer enough payday cycles to update their card. Past that, the recovery curve flattens sharply and you're paying gateway fees on declines that won't convert.
It's the foundation. Until involuntary churn is solved, you can't accurately measure the impact of any voluntary-churn intervention — your cancellation surveys, win-back offers, and pause flows all get contaminated by billing-failure noise. Dunning comes first in any retention roadmap.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.