---
title: "Food Cost Control for Catering: Cost per Head, GP% and Where the Margin Leaks | Havenue"
description: "How catering businesses actually control food cost: calculating cost per head, working out GP% before the quote goes out, the five places margin leaks on a banquet, and why a costing built last quarter is already wrong."
canonical: https://havenue.co/catering-food-cost-control/
markdown: https://havenue.co/catering-food-cost-control.md
language: en
last_updated: 2026-08-21
---
Resource &middot; Margin

# Food cost control for catering: cost per head, GP% and where the margin leaks

Last updated: August 2026 &middot; ~9 min read

Most caterers can tell you their food cost for last month. Very few can tell you the food cost of the wedding they quoted this morning. That gap is where catering margin is won and lost, because the price is agreed weeks before anybody knows what the food cost. This guide covers where catering margin actually leaks, how cost per head and GP% really work, and what to measure every week.

The short version

- Cost per head is the number the client argues about. GP% is the number that decides whether you stay open.
- A costing is only as current as the ingredient prices behind it.
- Yield is the gap between what you buy and what you serve, and it is where most of the loss hides.
- Quoted margin and realised margin are different numbers. Compare them per job, not per month.
- Weekly beats monthly. A monthly food cost tells you something went wrong four weeks ago.

## 1. The problems catering teams hit with food cost

- **The price is agreed before the cost is known.** A wedding quoted in March runs in September on September prices.
- **Costings live in a spreadsheet one person maintains.** When that person is busy, every quote goes out on old numbers.
- **Nobody re-costs after a supplier price move.** There are 400 recipes and one of them just got more expensive, and finding which is a day of work.
- **The quote and the actual never meet.** The job is done, the invoice goes out, and nobody compares the two.
- **Waste is invisible.** Trim, over-production and returns are real cost that never appears on any document.
- **The monthly number arrives too late.** By the time the food cost is reported, the jobs that caused it are finished and paid.

## 2. Cost per head is the number the client argues about

Catering sells per head, so cost per head is the number everything rests on. Get it right and every other decision is arithmetic. Get it wrong and you are negotiating from a number that was never true.

The usual mistake is counting only the main courses. A guest also eats the canape, the bread, the side, the sauce, the petit four and the coffee. Add a wastage allowance too, because a kitchen that produces exactly the covers will run short and everybody knows it.

Cost per head, live






Cost per head and GP% while the menu is still being built, not after the price is agreed.

## 3. GP% and how caterers actually calculate it

Gross profit is what is left of the selling price after the food. If you sell a menu at £60 a head and the food costs £18, the gross profit is £42 and the GP is 70%.

Two things trip teams up. First, VAT: GP is worked on the net price, not the price the client sees. Second, staff: labour is not in food GP, so a job with a healthy food GP can still lose money once the hours are counted. Read [labour costs in UK hospitality](/hospitality-labour-cost-control/) for that half of the sum.

**Worth knowing:** quote from GP%, not from a markup multiple. A 3x markup gives a 66.7% GP on one job and something else on the next, because the multiple ignores where the cost sits.

## 4. Where the margin leaks: five places

- **Old prices.** The costing is right for the day it was built and wrong every day after.
- **Yield.** The recipe says 220g of beef. The beef arrives whole and loses 30% to trim.
- **Over-production.** The prepared count is higher than the billed count, and the difference is food you paid for.
- **Uncharged extras.** The late canapes, the extra hour, the second round of coffee.
- **Substitutions.** The supplier sent a different brand at a different price and nobody re-costed the dish.

## 5. Yield, and the gap between the recipe and the kitchen

Yield is the difference between what you buy and what you serve. A whole side of salmon is not portions of salmon. A case of leeks is not trimmed leek.

A costing built on purchase weight rather than usable weight is wrong in the same direction every single time, which is what makes it dangerous: the error never cancels out. Cost every recipe on the yield your own kitchen gets, not the yield on the supplier’s sheet, and check it again when a supplier changes.

## 6. Supplier price moves, and last quarter’s costing

The single most common cause of a bad food cost month is a costing nobody updated. Prices move weekly and recipes do not.

The fix is not a better spreadsheet habit. The fix is making the supplier bill update the ingredient price, so every recipe that uses it re-costs on its own. See [supplier invoice scanning](/supplier-invoice-scanning-guide/) for how that link works.

Price moves, flagged






A price that moves should find the recipes, not wait for somebody to go looking.

## 7. Quoted margin against realised margin

Two numbers matter per job. The margin you quoted, and the margin you made. Most catering businesses only ever see the first one clearly.

Compare them job by job and the pattern shows up fast. It is usually one room, one menu or one type of booking that drags the average, and once you can name it you can price it differently or stop selling it.

Quoted against realised






Per job, not per month. The average hides the job that lost money.

## 8. What to measure weekly

- Food cost percentage for the week, against the same week last year
- Quoted GP against realised GP on every job that ran
- Ingredient prices that moved more than five percent
- Waste recorded, by reason
- Prepared count against billed count
- Extras served against extras charged

Weekly beats monthly for one reason. A monthly number tells you something went wrong four weeks ago, and by then the same thing has happened another four times.

## 9. What this means for your role

### Head of catering

You are judged on GP% across the book. What you actually need is the spread: which jobs, which rooms and which menus come in under the quoted margin, and why. An average hides every one of those.

### Sales director

You need to know, during the call, what a substitution does to the margin. Being able to say yes to the halibut and still hold your number is worth more than any discount you could offer instead.

### Executive chef

You own the recipes, so you own the costings. Yield percentages and portion weights are your levers, and they move the number more reliably than shopping around does.

### Head chef

You see the loss before anybody else: the trim, the over-portioning, the tray that came back. Recording it is what turns a feeling into a number somebody can act on.

## 10. Questions people ask

### How do you calculate cost per head for catering?

Add the ingredient cost of every dish a guest is served, including the garnish, the bread and the coffee, and divide by nothing: cost per head is per guest already. The mistake is stopping at the main courses. Canapes, sides, condiments and the wastage allowance belong in the number.

### What is a good GP% for catering?

Most UK caterers work to a food GP of 65% to 75%, meaning food cost is 25% to 35% of the food selling price. Banqueting at volume usually sits higher, private dining with expensive proteins usually sits lower. The useful comparison is your own quoted GP against your own realised GP, not somebody else’s benchmark.

### What is the difference between food cost percentage and GP%?

They are two views of the same split. If food cost is 30% of the selling price, the gross profit is 70%. Caterers usually talk in GP%, kitchens usually talk in food cost percentage, and the confusion between them costs people money in conversations.

### How often should recipe costings be updated?

Every time a supplier price changes, which for fresh produce and protein means weekly. Doing it by hand at that frequency is not realistic, which is why the price should flow into the recipe from the supplier bill.

### Why is my quoted margin better than my real margin?

Four usual reasons: the costing used old prices, the yield was worse than the recipe assumed, the prepared count was higher than the billed count, and extras were served but never charged. All four are measurable if the quote and the actual sit on the same record.

## 11. Where Havenue fits

Havenue costs the menu while it is being built, shows the cost per head and the GP% before the price is agreed, and re-costs every recipe when a supplier bill changes an ingredient price. The quoted margin and the realised margin sit on the same catering event record.

See [set menu creation](/features/set-menu-creation/), [ingredients, costs and suppliers](/features/ingredients-costs-suppliers/) and [reporting and compliance](/features/reporting-and-compliance/).

[Start free trial](/#pricing)

[See set menu costing](/features/set-menu-creation/)

**Read next**

[Supplier invoice scanning &rarr;](/supplier-invoice-scanning-guide/)

[Catering inventory and stock management &rarr;](/catering-inventory-management/)

[Labour costs in UK hospitality &rarr;](/hospitality-labour-cost-control/)

[&larr; Return to the Havenue home page](/)
