The method: five phases, one document between each

Five documents, one between each phase, each published as a sample so it can be read before it is commissioned.

The baseline reading is the examination's output: every measure the engagement will later be judged on, with the method that produced it and a confidence rating against each. Where two derivations disagree, both are shown. Where a figure is weak, it says so.

The findings and recommendation document names the constraint and the smallest set of changes that moves it, ranked by leverage. It is a go or no-go conversation, and one of the available answers is no.

The operating cadence sets what runs, who owns it, and what gets reported when.

The before-and-after measurement set re-runs the baseline on the same instruments, the same locations and the same definitions. Same method or the comparison is worthless.

The handoff packet is what your team keeps: definitions, instruments, access, and the revocation of ours.

All five are published as samples, redacted, with the redactions explained.

Because the alternative is buying a remedy for a condition nobody has measured, which is how most of the money in this industry is spent.

Three things go wrong without it, and all three are common. The first is treating the wrong constraint: acquisition gets funded while the loss is at the desk, and the group spends a year widening a funnel that leaks at the bottom. The second is that the definitions were never frozen, so a year later the before and after cannot be compared and everybody argues from a position they already held. The third is that nobody established what the group's own numbers were, so performance gets judged against an industry figure that was published by somebody with a product to sell.

Six weeks is the price of not making those three mistakes, and it is a fixed fee rather than a retainer for exactly that reason. If the examination finds nothing worth fixing, we say so and we leave — which is the outcome that makes the rest of it honest.