In this guide 6 sections
Food cost differs because the same recipe is not cooked, portioned, wasted or recorded the same way at every outlet. Compare theoretical usage, worked out from recipes and sales, against actual stock movement per outlet. The gap then points to transfers, unlogged wastage, portion drift or menu items with no recipe.
Start with theoretical against actual usage
Theoretical usage is what the kitchen should have consumed. Take each item sold at the point of sale, multiply by its recipe, and add up paneer, chicken, oil, flour and packaging across the day. That is the expected figure.
Actual usage is opening stock plus purchases and transfers in, minus transfers out and closing stock. One is a calculation, the other is a count. The difference between them is the variance, and it is the only number worth chasing.
Compare it outlet by outlet, ingredient by ingredient, week by week. A single brand-wide food cost hides everything. One outlet can be clean while another leaks, and the average looks fine.
Central kitchen transfers move cost around
Many groups prepare gravies, marinades, doughs and sauces at a central kitchen and send them to outlets. If the transfer note carries no quantity, or carries a rough one, the receiving outlet books cost it did not incur.
The reverse happens too. A dispatch recorded at the central kitchen but never received on paper leaves stock sitting on one ledger and cooked on another. Both sides then show strange variance that cancels out only at group level.
Weigh or count at both ends. Record the transfer once, as an outward note that the outlet confirms with its own received quantity and any shortage.
- Item, batch and prepared date on every transfer note
- Sent quantity and received quantity side by side
- Rejected or spoiled items logged at receipt
- Transfer price fixed so margin is not hidden
Wastage, portions and menu mapping hide in plain sight
Unlogged wastage is the quiet one. Burnt batches, trimmings, expired cream, returned plates and staff meals all leave stock without a sale. If nobody writes them down, they show up as variance and get blamed on theft.
Portion drift is the slow one. A new cook ladles a little more gravy, the fryer person cuts thicker chips, a cashier adds extra toppings. No single plate looks wrong. Across two hundred plates a day, it is a real cost.
Then there are unmapped items. A combo added at one outlet, a seasonal special, a modifier like extra cheese. If the point of sale item has no recipe behind it, its ingredients are consumed but never counted in theoretical usage.
- Wastage reasons chosen from a short list, not free text
- Standard portion weight per dish, checked by spot weighing
- Every sellable item and modifier linked to a recipe
- Staff meals and tastings recorded as their own line
What a custom stock build can flag by outlet
Off-the-shelf billing tools stop at sales. The stock side needs your recipes, your central kitchen routine and your outlet formats, whether dine-in, kiosk or delivery-only. A custom build can join all three.
The useful output is an exception list, not a dashboard of totals. It names the outlet, the ingredient and the likely cause: variance rising after a menu launch, transfers short two weeks running, wastage entries falling to zero.
Flags should reach the area manager within a day of the count, while the bin can still be checked and the cook can still be asked.
- Variance per ingredient per outlet against a tolerance
- Items sold with no recipe attached
- Transfers where sent and received quantities disagree
- Outlets logging almost no wastage
Run a two-week test at your worst outlet
Pick the outlet with the highest food cost and ten ingredients that make up most of your spend. Count them every morning and night for two weeks. Do not announce it as an audit; call it a stock trial.
Build theoretical usage by hand for those ten items from the sales export and your recipe cards. Put the two columns side by side. The first pattern you see, usually a transfer gap or an unmapped item, is your starting point.
Write down which recipes are missing from the point of sale. Fix those first, because every later comparison depends on them.
Outlet food cost variance worksheet
Every ingredient gets a row on this single sheet. Columns run left to right: opening stock, purchases, transfers in, transfers out, closing stock, actual usage, theoretical usage from recipes and sales, variance and a cause code. Cause codes are transfer, wastage, portion, unmapped item and count error. A header block holds the outlet, week and counter name.
Open a blank worksheet to printQuestions owners ask
Is higher food cost at one outlet always theft?
No. Most variance comes from paperwork gaps: transfers without quantities, wastage nobody logged, menu items with no recipe. Check those first. Treat theft as a conclusion after the records are clean, not a starting guess.
How often should each outlet count stock?
Count high-value and fast-moving items such as meat, dairy, oil and packaging daily or at least twice a week. Count dry goods weekly. A full count monthly keeps the books honest.
Do we need recipe costing before we can compare outlets?
You need recipe quantities, not prices. Quantity per dish is enough to work out theoretical usage. Prices can be added later to turn the variance into money.
Why does variance look fine at group level but bad at one outlet?
Group totals let one outlet’s surplus cancel another’s shortage, especially when central kitchen transfers are loosely recorded. Always read variance per outlet before adding anything up.