Vendor pages score themselves on their best month. Your rubric should score them on your worst one — the month the card file aged, a pricing change annoyed people, and failures arrived in a pile instead of a trickle. I evaluate payment recovery software on five tests, all of them observable before signing anything, and none of them requiring a sales call to verify.
Test one: classification before action. Ask what the tool does in the first hour after a decline, for each major decline family. Expired and replaced cards should trigger a customer prompt, not a blind retry. Insufficient funds should trigger spacing, not volume. Hard declines should trigger a short, dignified exit. If the answer is one retry schedule for everything, the tool is a timer with a logo, and your current platform settings already own a timer.
Test two: the card-update moment. Recovery is won or lost on the page where a customer fixes their card. Is it reachable from the email without a password reset expedition? Does saving a new card immediately re-present the failed charge, or does the customer fix the card and then wait for a nightly job while access stays suspended? That gap — card saved, charge still pending — is where goodwill goes to die, and it is entirely checkable in a demo.
Test three: counting you can audit. Every recovered dollar should carry a cause: our retry, our message, the customer's own update, the platform's native attempt. Software that reports one blended recovery number is asking to be credited for weather it did not make. Blended numbers also hide the guarantee question — if a recovery promise exists, the counting method behind it should be written down in the terms, not described warmly on a call.
Test four: price behavior at success. A percentage-of-recovery price feels aligned until you model it at twice your current volume, where it quietly becomes the most expensive line in the stack. A flat price is boring in exactly the way a utility bill is boring. Score this test with a spreadsheet, not a feeling: your recovery today, recovery improved by half, recovery doubled. Three rows, two pricing models, one obvious answer for most books.
Test five: the handover from your platform. Your billing platform already retries on its own schedule, with its own logic. Recovery software earns its place in the gaps — and only if the handover is explicit. Who retries first? What stops the two systems from attempting the same card on the same morning, which reads to the bank like desperation and to the customer like chaos? On some platforms the merchant must deliberately hand failed-payment handling over before an outside tool's attempts take precedence; on others the weave is automatic. Either way, the answer should be a sentence you can read in the documentation, not a reassurance you received on a call. Double attempts are the signature of a stack nobody designed, and the bad month is when they multiply.
Test six: exit and silence. When a customer cancels mid-recovery, does outreach stop that minute? When a charge is refunded or disputed, does the recovery count correct itself? Software that keeps emailing a cancelled customer is not being thorough; it is manufacturing complaints and chargebacks. The bad-month version of this test is the one that matters, because bad months are when cancellations cluster.
Run those five tests and most of the category sorts itself. What survives is recovery software that treats every failed renewal as a case to work, not a queue to blast — classify first, prompt clearly, count honestly, price flat, and stop the moment the customer says stop. The features lists will keep growing on every vendor's page. The bad-month behavior is the product.
Published by Swipe Again, the flat-fee failed payment recovery service for subscription businesses.
Top comments (0)