Restaurant margins are thin enough that a two-point move in food cost is the difference between a good year and a worrying one. The frustrating part is that the information needed to catch that move is already sitting in your system - it is just not being looked at in the right shape.
1. Food cost percentage, weekly not monthly
Calculating food cost monthly means you find out about a problem an average of six weeks after it started. Weekly is the minimum useful cadence: opening stock plus purchases minus closing stock, over sales, for the week. It takes a cycle count of your top-value ingredients, not a full inventory.
2. Theoretical versus actual usage
This is the single most valuable number in a kitchen and the one most restaurants never calculate. Theoretical usage is what your recipes say you should have used, given what you sold. Actual usage is what disappeared from the store. The gap is over-portioning, wastage, staff meals or theft - and the gap by ingredient tells you exactly where to look.
3. Plate cost, kept current
A recipe costed at the price of ingredients eighteen months ago is fiction. Costs must recalculate when purchase prices change, which means recipes need to reference live ingredient costs rather than a number someone typed into a spreadsheet in a previous season. Set a margin alert so that when an ingredient price moves enough to push a dish below your target margin, someone is told before the next hundred plates go out.
4. Menu engineering: the four boxes
Plot every dish on two axes - popularity and contribution margin - and you get four groups that each need a different action.
| Group | Profile | Action |
|---|---|---|
| Stars | Popular, high margin | Protect fiercely. Never cut, never cheapen, feature prominently. |
| Plough-horses | Popular, low margin | Re-engineer the recipe or nudge the price. Small changes, big total effect. |
| Puzzles | Unpopular, high margin | Reposition, rename, describe better, or have servers recommend it. |
| Dogs | Unpopular, low margin | Remove. They cost you prep, storage, waste and menu space. |
5. Wastage, recorded without blame
If recording waste feels like a confession, nobody records it, and it reappears as an unexplained variance. Make it a two-tap action with reason codes - spoiled, dropped, over-produced, comped, staff meal - and treat the resulting report as information about your process rather than evidence against your staff.
6. Yield, honestly measured
A whole chicken does not produce a whole chicken's worth of sellable portions. Neither does a case of tomatoes. If your recipes cost inputs at purchase weight rather than usable yield, every dish is under-costed and your theoretical usage is wrong. Measure yield once per major ingredient and let the system apply it thereafter.
7. Purchase price variance
Suppliers raise prices quietly. A report showing what you paid for each ingredient over the last twelve months, with the percentage change, is the single most useful document to have in front of you at a supplier negotiation - and it takes seconds to produce if your purchasing runs through the same system as your kitchen.
8. Prime cost
Food cost alone is only half the picture. Prime cost - food plus labour as a percentage of sales - is the number experienced operators actually watch, because the two trade off against each other. A cheaper ingredient that takes twice as long to prepare has not saved you anything. Track them together, weekly, per outlet.
None of this requires a consultant or a new spreadsheet. It requires recipes in the system, purchases going through the system, wastage recorded, and someone opening the same four reports every Monday morning. The restaurants that do this consistently are not smarter than the ones that do not. They are just looking.
Want this handled by software rather than a spreadsheet?
SumPOS does most of what this article describes automatically - and the free plan is genuinely free, so you can test the idea before committing to anything.