By Ryan Richardson · Published 8 October 2026
Once the order value problem had been correctly diagnosed as a pricing issue rather than a traffic issue, the fix itself still wasn't a single clean move. It took three goes, spaced months apart, to actually land.
The first diagnosis was right: the problem was price, not traffic. But the first and second repricing attempts underpriced the fix again, in a smaller way. Each round moved the price, read the next week's settled order value, and found it still short of the break-even floor. Only the third move cleared it.
Diagnosing the right problem and correctly solving it on the first attempt are two different skills, and the first doesn't guarantee the second. Pricing a digital ladder against a floor built from estimates, rather than from fifty or more settled purchases, produces a guess with a decimal point attached. Two of those guesses, in a row, still landed short.
Ladder review now triggers automatically whenever order value moves, rather than waiting for a manual check. Each reprice gets read against settled cash the following week, not against a projection, and a repricing that doesn't clear the floor gets treated as an incomplete fix rather than a finished one.
After any price change, read the next full week's settled order value against your break-even floor before declaring the fix done. If the order value moved in the right direction but is still short, treat that as signal to move again, not as evidence the fix doesn't work. Expect acquisition cost and order value to take roughly fifty purchases to settle; judging a reprice on ten sales is reading noise.
| Claim | Value | Source |
|---|---|---|
| Number of repricing attempts before the funnel cleared its floor | three | Measured in Real Money, Field Manual |
| Time between repricing attempts | months apart | Measured in Real Money, Field Manual |
| Purchases needed before an acquisition cost or order value figure settles | roughly fifty | THE BOOK FULL.md Piece 15, line 1737 |