A restaurant's profit is decided in the back office: supplier prices that move unnoticed, plate cost never recalculated, and counts that never happen. What to check weekly.
Key takeaways
- Ingredient prices move continuously, and a menu that is not updated absorbs the difference.
- A plate cost never recalculated is a stale assumption, not a figure.
- A weekly count of a few high-value items is worth more than a full monthly one.
- Rostering to forecast traffic rather than to availability is the biggest lever on labour cost.
- Waste that is not recorded does not exist in the reports, and so is never addressed.
In a restaurant, profit is decided in the back office rather than in the dining room. A menu priced against last year's ingredient costs, a stock count done "roughly", and a rota built around who is available - all three erode percentage points, and all three are fixable without changing anything in the kitchen.
Supplier prices: the quiet leak
A restaurant buys from ten to twenty suppliers, and prices move - seasonality, market prices, and sometimes simply an increase nobody announced. When nobody compares invoice to invoice, a price that rose thirty percent passes unnoticed.
The check that catches it is not complex: a list of ten to fifteen items making up most of the spend, and tracking their unit price over time. Unit price rather than invoice total - because a pack that shrank is a price rise that does not look like a price rise.
What adds immediate value is checking the other direction too: an item ordered and delivered in a quantity different from the one billed. That is a common error, and it almost never surfaces without comparing against the delivery note.
Plate cost: why it goes stale
| The element | How it moves | The effect |
|---|---|---|
| Ingredient price | Seasonal and market | Direct, and usually unnoticed |
| Actual portion size | Drift from the recipe | Larger than people expect |
| Trim and preparation | Peeling, cleaning, offcuts | Usually left out of the calculation |
| Extras and sauces | Given freely | Accumulates |
| Waste | Plates returned or burned | Not in the calculation at all |
The second row is the big one. A recipe saying 180 grams served at 220 grams is twenty percent extra cost on the same dish, and it shows up in no report. Weighing a few plates a week reveals it immediately.
Counting: fewer items, more often
A full stock count is hours of work, so it gets postponed - and therefore happens monthly, sometimes less. The result is that when a gap appears, there is no way to know when it arose.
What works better is a weekly count of ten to fifteen high-value items only: meat, fish, alcohol, cheese. Those are both the items where a gap costs money and the items that can be counted in twenty minutes.
That count produces one number available no other way: how much was actually consumed against how much should have been consumed given sales. The gap between them is waste, theft, or oversized portions - and without the count it simply does not exist.
Shifts: rostering to traffic
At most restaurants the rota is built around who is available and who asked. That is fair and it is also disconnected from demand: some shifts carry two people too many and some are one short, and both cost money - one in labour, the other in service.
What changes this is a simple figure every till already holds: sales by hour, by day of the week. Four weeks of data give a stable pattern, and the rota is built against it.
It is also what lets you answer the real question - what an hour of labour costs relative to sales in that hour - and that is the number the hard decisions come from.
Waste: if it is not recorded, it does not exist
A burned dish, a product that expired, an order mistake that came back - all cost money and appear nowhere. The result is that actual food cost is higher than the calculated one, and nobody knows why.
The fix is one sheet in the kitchen with three columns: what, how much, and why. Five seconds per item. After a month you have a ranked list of causes, and usually the top two explain most of the amount - and both are fixable.
What to check every week
- Unit prices of the key items, against last week.
- A count of the high-value items.
- Total recorded waste, by cause.
- Labour cost as a percentage of sales, by day.
The fourth check is the one showing whether the rota works. A day where the percentage jumps is not necessarily a bad day - sometimes it is a day where people were rostered by availability rather than by traffic, and that is exactly what can be corrected next week.
Sources
Frequently asked questions
Do you need restaurant management software?
Not to start. The four checks above work in a spreadsheet, and once it is clear which of them takes the most manual time - that is the point at which a system pays for itself.
How often should a menu be updated?
On price, at least once a season or when a key cost moves materially. More important than the frequency is that the update rests on a recalculated plate cost rather than on a feeling.
What do you do when a supplier raises a price?
First confirm it actually happened, against unit price rather than invoice total. After that it is a commercial conversation, and it looks entirely different when you arrive with a figure rather than an impression.
Should small waste be recorded too?
Yes, because the value is in the series rather than the item. Small recurring items are exactly what accumulates, and they are also what explains a gap in the count that otherwise looks inexplicable.
Keep reading
Related service
Inventory & Purchasing
SKU-level stock, reorder rules and an approval flow that leaves a record.
About the author
Yehonatan Saadia
Freelance automation, web & MVP developer
I'm Yehonatan Saadia, a senior developer who builds business automation, custom websites, and MVPs for small and mid-sized companies across the US, Europe, and Israel. These guides come from real client work, not theory.
Work with meHave a project like this?
Tell me what you're trying to automate or build and I'll tell you the fastest reliable way to ship it.