Ask three subscription founders what failed payment recovery software does and you will get three different answers, all of them confident. One describes a retry engine: software that re-attempts a declined card on a smarter schedule. One describes a messaging tool: emails and texts that ask the customer to update a card. One describes a full recovery operation: detection, retries, outreach, card-update pages, and reporting that separates what recovered on its own from what needed work. They are not disagreeing about quality. They are describing three different products that happen to share a shelf label.
That confusion is expensive because teams buy the first product while expecting the third. A retry engine is genuinely useful. Most first failures are soft declines — temporary states like insufficient funds at that hour, a processor hiccup, a velocity check — and a well-timed second attempt collects a real share of them without troubling the customer at all. But a retry engine cannot help the customer whose card was replaced, whose bank now wants authentication, or who never learned the charge failed. Those renewals do not need another attempt on the same credential. They need a person to act, and a product that never asks the person to act will quietly lose every one of them.
The messaging-only product has the opposite shape. It asks the customer to act, sometimes well, but it treats every failure as a communication problem. For the large pile of soft declines that would have cleared on a Tuesday retry, an urgent please-update-your-card email is noise at best and a cancellation prompt at worst: you have told a happy customer, in writing, that something is wrong with their payment, and offered them a natural exit point in the same breath. Outreach earns its place after retries have had their chance, or immediately for the failure classes where retries are pointless — not as the opening move for everything.
This is why a category read that separates retry engines from full recovery systems matters before any demo is booked. The useful evaluation question is not which product has the longest feature list. It is which failure classes each product actually covers, in which order, and what happens to the classes it does not cover. Put the four piles on the table — retryable soft declines, replaced or expired credentials, authentication-required failures, and customers who simply need a clear notice and a working update link — and make every vendor walk a renewal through each pile. The gaps appear within minutes, in the vendor's own words.
Pricing hides in the same fog. A percentage of recovered revenue sounds aligned until you notice it charges most for the easiest pile, the soft declines that a competent retry schedule would have collected anyway. A flat monthly fee sounds blunt until you price the alternative honestly: what does one retained subscriber pay over the following year, and how many of them does the fee actually require each month to break even? Neither model is virtuous on its own. The arithmetic only becomes visible once you know which piles the product works, because a fee charged on the baseline pile is a tax, while the same fee charged on the pile you were losing entirely is a bargain.
There is a practical test any team can run this week without buying anything. Take last month's failed renewals and sort them by decline type and final outcome. Count how many recovered with no customer action, how many recovered after the customer acted, and how many were lost while the system was still politely retrying a dead credential. That third count is the size of the problem a full recovery system exists to solve, and it is almost never the number on the dashboard's front page.
Swipe Again is built as the third kind of product: retries woven around the platform's own schedule, outreach reserved for the failures that need a person, and reporting that keeps the piles separate so the fee can be judged against the middle work. Whichever product a team chooses, the category stops being confusing the moment the piles are named. Buy for the pile you are actually losing.
Top comments (0)