Dunning pricing hides in two places: in the percentage a vendor takes from whatever it recovers, and in the tier ladder that charges more as the store's revenue grows. Both shapes share a property worth naming — the store cannot know its annual cost on the day it signs, because the bill depends on outcomes and growth that have not happened yet. Flat-fee dunning removes that unknown, and the removal itself is the feature: one number, known in advance, against which the recovery can be judged without a calculator fight.
The arithmetic a store should run first is deliberately unflattering to everyone. Start with the workable pool: failures still open after the platform's native retries, belonging to customers who never asked to leave. Not the raw failure count, which includes losses no tool could touch, and not the vendor's favourite subset either. Multiply the pool by a recovery share at the pessimistic end of the published range, not the headline end. Whatever survives that multiplication is the honest monthly value of the work. A store can read how flat-fee dunning is priced and what the single number has to cover and then hold its own figure against the fee in one line: does the pessimistic value clear the cost with room, or does the case only work at the optimistic end?
Percentage pricing scrambles that test in a specific way. When the vendor takes a share of recovered revenue, the store's cost rises exactly when recovery rises, which sounds fair until the store notices it is also paying the share on recoveries its own platform would have produced unaided, unless the attribution is stamped carefully at the moment of each save. The question to ask any percentage vendor is not the rate; it is the denominator and the stamp. Tiered pricing scrambles the test differently: success pushes the store into a more expensive band, so the tool's price grows with the store's revenue whether or not the failure pool grew with it.
None of this makes flat pricing automatically honest. A flat fee can still be wasted money on a pool too small to justify it, which is why the guarantee matters more than the shape: a defined recovery floor over a defined window, judged on total recovery, with a consequence the vendor actually pays when it misses. The shape tells the store what it will spend; the guarantee tells it what happens if the spending does not work. A vendor offering the first without the second is asking to be trusted on exactly the point the arithmetic cannot settle in advance.
One more test belongs in the sequence, and it is about time rather than money. Ask when the fee starts relative to the work. A percentage vendor begins costing the store only when recoveries land, which flatters cash flow in month one and bills the store forever after. A flat fee starts on a date, trial first, and its fairness depends on the recovery arriving inside the same window the guarantee measures. Neither shape is a trick; they are different allocations of risk across the first quarter. A store with a large, old failure backlog should weigh that allocation honestly, because the backlog is where flat pricing is usually cheapest and percentage pricing most expensive.
Run the pessimistic test first, then read the guarantee, then argue about features. That order protects the store from the most common buying error in this category, which is falling for a dashboard before checking whether the fee survives contact with an unflattering denominator.
Swipe Again publishes its number plainly: a flat $179 a month for the full recovery work, with the guarantee judged on total recovery across the first ninety days. If a store's pessimistic arithmetic does not clear that fee, the honest answer is to wait until the pool grows. Pricing in the open means accepting that answer too.
Top comments (0)