Every recovery vendor's homepage carries a number: dollars recovered, subscribers saved, revenue rescued. Almost none of them tell you the counting rule behind the number, and the counting rule is where the honesty lives. A recovered dollar can be defined so loosely — any subscriber who paid anything after any failure, ever — that the metric measures the passage of time rather than anyone's work.
Here is the strict version we hold ourselves to, offered as a rubric you can aim at any vendor, including us. A recovery counts when three things are true. First, there was a real failure: an actual declined charge on an actual subscription, not a card that was merely about to expire. Second, the recovery is stamped at the moment it happens — the successful charge — with the attribution recorded then, not reconstructed at month-end from a spreadsheet and a hopeful memory. Third, it is counted through exactly one door: if two systems touched the failure — your billing platform's native retries and a recovery tool working around them — the dollar is credited by a rule decided in advance, never by both, and never by whichever report is being shown to a prospect.
That third rule is the one the industry resists, because double counting is so useful. Stripe's own retries recover a real share of failures; so do Recharge's and Chargebee's. A vendor who weaves additional attempts into the gaps of that native schedule is doing genuine work — and then, too often, claiming the native schedule's recoveries as well. The honest separation is not technically hard. It requires keeping, for every failure, a single record of what was attempted, by whom, and what finally succeeded. It is bookkeeping, and bookkeeping is unglamorous, which is precisely why it is a good test: a company careless with its own numbers will be careless with yours.
There are softer forms of inflation to watch for as well. Counting a save at full contract value when only one payment was recovered. Counting a subscriber who cancelled a week later as retained. Reporting gross recovered revenue with no line for what the recovery itself cost — the percentage fees, the discounts offered in save flows, the support time. A number that cannot survive those questions is marketing, and it should be priced as marketing: worth nothing in a buying decision.
We published our own rule at how Swipe Again counts recovered revenue before we had numbers worth bragging about, on the theory that the rule has to predate the results or it will quietly bend to fit them. If you are evaluating recovery tooling, ask every vendor for the equivalent page. The ones who have it will show you. The ones who don't have just told you something worth more than their headline number.
The recovery itself matters, of course — timing, decline-aware retries, emails a person would actually act on. But in a category where everyone sells the same noun, the counting is the character. Swipe Again's flat $179 recovery service is $179 a month flat partly so the arithmetic stays auditable: one price, one counting rule, no percentage of a number we defined ourselves.
Top comments (0)