Decline recovery

Decline recovery is the practice of turning card payments an issuer refused into approved ones through better-formed requests, repaired credentials and well-timed second attempts, and for recurring or high-volume card businesses it functions as a revenue line. Response codes split failures into soft declines, such as insufficient funds or velocity limits that often clear hours later, and hard declines, such as closed or stolen cards, which should never be retried. Account updater services and network tokens fix stale credentials before renewals fail. Full billing data, correct merchant category, proper recurring flags and SCA exemptions lift first-attempt approvals. Because recovered payments carry no acquisition cost, one percentage point of approval rate on 2 million EUR of monthly volume adds 20,000 EUR.

16 October 20252 min read

Decline recovery is the practice of turning card payments an issuer refused into approved ones through better-formed requests, repaired credentials and well-timed second attempts, and for recurring or high-volume card businesses it functions as a revenue line. Response codes split failures into soft declines, such as insufficient funds or velocity limits that often clear hours later, and hard declines, such as closed or stolen cards, which should never be retried. Account updater services and network tokens fix stale credentials before renewals fail. Full billing data, correct merchant category, proper recurring flags and SCA exemptions lift first-attempt approvals. Because recovered payments carry no acquisition cost, one percentage point of approval rate on 2 million EUR of monthly volume adds 20,000 EUR.

Not every decline is worth chasing

Response codes split failures into two groups, and that split governs the whole program. Soft declines are temporary, covering insufficient funds, issuer velocity limits, timeouts and fraud flags that clear, and the same card often works hours later. Hard declines are final: closed account, card reported stolen, invalid number. Retrying those earns fees and scheme attention and nothing else.

A program that ignores the code and retries everything drags the merchant's authorization rate down, which acquirers monitor. Deciding which soft declines get a second attempt, and when, is the job of retry logic.

Repairing the card instead of resending the request

A large block of renewal failures is not a payment problem but a stale credential. Card networks run account updater services that refresh reissued and expired numbers held on file, and network tokens do the same job continuously, since the token survives a reissue. Both remove failures before they occur, which is cheaper than recovering them afterward.

Where the credential genuinely cannot be repaired, no amount of resubmission helps and the only route is the customer.

Declines you can prevent at submission

First-attempt approvals rise with request quality. Send full billing and address data, use the correct merchant category, flag the transaction properly as recurring or customer-initiated, and apply an SCA exemption where the payment qualifies instead of pushing every customer through a challenge. Routing matters too, because acquirers hold different issuer relationships and the same traffic approves at different rates through each. Much of this belongs to checkout optimization rather than to billing.

Asking the customer without sounding like spam

When the card itself has to change, the message decides the outcome. A note naming the product, the amount, the date and the last four digits, with a direct link to update the payment method, recovers far more than a generic dunning sequence. Send it from an address the customer recognizes, and stop sending the moment they act.

What one point of approval rate is worth

Recovered payments carry no acquisition cost, so almost all of the value drops through. A business processing 2 million EUR a month gains 20,000 EUR of monthly volume from a single percentage point, with no new customer, campaign or discount behind it. That is why approval rate belongs in the same review as conversion rate, and why it usually beats a front-end experiment for effort spent.

Comments

Related terms