Five stages, a fixed fee agreed before we start, and a number we both accept as the measure of whether it worked.
One paid day, on your busiest service. Two of us: one in the kitchen with a timer, one in the dining room counting what actually leaves the pass and what comes back. No interviews, no workshop, no questionnaire. At the end you get four pages: what we saw, what it is probably costing, and whether an engagement is worth your money.
Roughly one visit in six ends with us saying no engagement is needed. That is the point of charging for the day — it lets us say so without having lost anything.
Before any engagement starts we agree, in writing, the single measure it will be judged on. Food cost percentage. Ticket time at the pass on a Friday. Prep hours per hundred covers. Waste weight per week. One number, chosen by you from the ones we can actually move.
Two to four weeks of counting, depending on the engagement. Invoices, POS mix, weighed waste, timed tickets by station, and a walk of every cooler and dry store. We take our own measurements rather than accepting the ones already in your system, because the system usually records what somebody intended rather than what happened.
| What we measure | How | Typical duration |
|---|---|---|
| True plate cost | Purveyor invoices, last four periods, matched to recipe yields | 1 week |
| Sales mix | POS export by daypart, cleaned of comps and staff meals | 2 days |
| Waste | Weighed, split prep / spoilage / re-fire | 5 days |
| Ticket flow | Timed by station across 3–5 services | 1–2 weeks |
| Labor against demand | Scheduled hours vs covers by half-hour | 4 weeks of history |
Everything we recommend is sequenced, because a kitchen cannot absorb fourteen changes at once and the ones made in week one determine whether the rest are possible. Removals first, then re-costing, then layout, then price. Menu design changes last, once the plates behind them are stable.
We work the changes with your chef, on your line, during service. A recommendation that has never been executed under pressure is a hypothesis.
Handover is a working session, not a document. Your team maintains the recipe cards, the par sheets and the waste log afterwards; if they cannot, we built the wrong system and we fix it before we leave.
Ninety days later we come back for a day and re-measure the agreed number, at no charge. That visit is in every contract. It is also the reason we decline work where the operator will not commit to keeping the measurement going after we leave — without it, nobody can tell whether the engagement worked or the weather simply improved.