Food Cost in Catering Tenders: Why the Percentage on Paper Is Not the Cost You Will Live With

Every catering tender has a moment where the buyer looks at the food cost column and feels in control: a product list, neatly priced, a percentage at the bottom. It looks precise, comparable, easy to score.

And after more than 100 tenders, I can tell you that it is one of the least reliable numbers in the whole file. Not because caterers lie, but because the number is theoretical by design. The food cost you price in a tender is not the food cost you will live with. What you will actually live with is decided in the kitchen, every day, long after the contract is signed.

This post is for the facility, procurement and workplace teams who write and score those tenders: where the gap comes from, what it looks like in numbers, and what to ask for instead.

What food cost is, and where it sits in the money

Food cost, or cost of goods, is simply the cost of the raw ingredients that go into what you serve. In a typical workplace restaurant it represents 28 to 35% of total revenue — next to manpower at 45 to 55%, other costs at 8 to 15% and a gross margin of 0 to 19%.

Because food is the part of catering you can taste, it is where clients instinctively push: "bring your food cost down." But a 30% food cost can be excellent or terrible depending on how it got there. The percentage is the result, not the lever.

The theoretical approach: how tenders treat food cost

When you go to market, the bidders are almost always handed a product list and asked to price the food cost on it. It looks like hard information. In practice it carries very little workable information, because an item's food cost is purely theoretical.

One dish can be made from many different recipes, built cheaply or richly, with a food cost that ranges from simple to double. So when five caterers price the same list, you are not comparing five food costs. You are comparing five unwritten assumptions about recipes, portions and attendance.

The tender also treats demand as flat: average attendance times average dish cost. That fills a spreadsheet. It is not how a restaurant behaves.

The reality: food cost is set by daily operational management

In operation, food cost is mainly driven by day-to-day management, not by the price you pay the supplier. The supplier price is the part everyone looks at and the smallest part you can actually change. Four operational drivers do most of the damage.

Over-production — cooking more than the attendance needs, so the surplus goes in the bin. Rules on late availability — the promise that the full offer is still there at the end of service guarantees waste unless it is deliberately managed. Portions not respected — a serving spoon that is consistently too generous quietly moves your food cost by whole points. And menu management — which dish you choose to run on which day.

Behind those drivers sit five levers I keep coming back to: the offer (a sprawling offer buys a sprawling ingredient list), variation (every menu change opens new ingredients and leaves yesterday's half-used box at risk), waste (food you paid for and threw away — the purest loss in the operation), recipes (how dishes are engineered to share components) and flexibility (turning today's roast into tomorrow's component rather than tomorrow's bin). None of these are purchasing decisions. Waste in particular is overwhelmingly a design problem, not a discipline problem. You cannot train your way out of a menu that is built to waste.

A worked example: same dishes, same "theory", different food cost

Menu management is the driver buyers most underestimate, so let me make it concrete with my own food cost simulation.

Take three days with attendance of 80, 100 and 60 passages — an average of 80. Take three dishes with food costs of 2, 3 and 4 (an economic, a mid-range and an expensive dish) — an average of 3. The tender logic multiplies the averages: 80 × 3 gives a theoretical food cost of 240 per day.

But that figure only holds if every dish meets an identical crowd, which never happens. In reality, someone decides which dish runs on which day. Here are all six possible schedules:

Scenario	Day 1 (80)	Day 2 (100)	Day 3 (60)	Real food cost (avg/day)	vs theoretical (240)
3	Dish 2 (3)	Dish 1 (2)	Dish 3 (4)	226.7	−5.6%
1	Dish 1 (2)	Dish 2 (3)	Dish 3 (4)	233.3	−2.8%
5	Dish 3 (4)	Dish 1 (2)	Dish 2 (3)	233.3	−2.8%
2	Dish 1 (2)	Dish 3 (4)	Dish 2 (3)	246.7	+2.8%
6	Dish 3 (4)	Dish 2 (3)	Dish 1 (2)	246.7	+2.8%
4	Dish 2 (3)	Dish 3 (4)	Dish 1 (2)	253.3	+5.6%

Put the economic dish on the busiest day and the expensive one on the quietest, and real food cost lands about 5.6% below theory. Do the opposite and it lands about 5.6% above. Same three dishes, same theoretical food cost, a swing of roughly 11% between best and worst — created by nothing but the order in which they were scheduled.

And this example isolates a single driver — no waste, no over-production, no generous portions. None of it shows up on the product list you asked bidders to price.

The production case: where food cost and manpower meet

The second illustration is the production case from the ALOHA material. Picture a classic week with 15 recipes, averaging 10% waste, needing around 30 production hours. Redesign it to 7 recipes with shared components and smarter batches, and waste falls to about 4.6% with roughly 18 hours of production. Push to 3 well-engineered recipes and you approach 2% waste and about 12 hours.

As recipes go down, waste and production hours go down together. Nobody bought cheaper food and nobody was paid less. This is why food cost and manpower can never be managed in separate meetings: they are the same decision seen twice. And it is why "more choice, less variation" beats "more variety" — you can still give eaters plenty to choose from while stabilising the production behind it.

One honest caveat: cut variation too hard and quality perception drops, participation falls and the subsidy rises to compensate. Match the offer to the people actually eating, then stabilise production behind it.

Why the contract determines whether you ever see real food cost

Your contract structure decides who feels the gap between theory and reality. In an open book contract you pay actual costs plus an agreed fee: if food cost comes in at 31%, you pay 31%. You get full transparency, but you have to supply the pressure, because the caterer's incentive to drive costs down is weaker. In a single forfait, the caterer carries the risk — and prices in a premium for it — and once the price is fixed, their margin improves every time quality quietly drops. A multi-forfait typically prices food per item sold, so it moves with what people actually eat, but it only works if the forfaits are set honestly and the consumption data exists.

Then there is the subsidy, which in most contracts is not negotiated at all; it is the residual you discover once the year has run. That makes it the perfect hiding place for food-cost problems. Over-wide menus and excessive waste do not show up on the plate; they show up as a larger subsidy that nobody decided on.

What buyers should do instead

Never tender on food-cost percentage alone. Tender on the design that produces good food cost. Ask bidders how many recipes they plan to run per week, how they engineer shared components, what waste level they commit to and how they will measure it, and how they build in flexibility to re-use. Ask them to show you their version of the production case — and how they plan the weekly menu against your attendance pattern.

That last point means you have to give them an attendance pattern. Too many tenders are a price exercise with too little data, too little time and too little freedom for the caterer to analyse real needs. Roughly 80% of the time, an underperforming contract was lost in the preparation, not in the kitchen. Measure your real passages, by day, before you go to market.

Quick checklist for your next tender

• Ask bidders how they will manage food cost — recipes, shared components, waste target, re-use — not just to price a list.
• Share real attendance data by day so menus can be planned against it.
• Require a measured waste commitment and reporting method.
• Choose a contract that makes real food cost visible — and assign someone to read the numbers.
• Track subsidy per meal so food-cost drift cannot hide inside it.

A caterer who can talk fluently about recipes and waste is managing food cost at the source. One who only quotes you a percentage is managing your perception.

---

Food cost is Session 5 of the ALOHA Catering Masterclass, a 12-session programme for the people who decide, buy, tender and govern workplace catering — covering the money map, manpower, contracts, subsidies and consumer profiles. To learn more about the Masterclass, or to talk about your next tender, visit aloha-consultancy.com.