The labour cost hidden in your menu
Two dishes can return the same gross profit and cost you completely different amounts to put out. One of them is on your menu because the GP looked fine. This is how to find it, and why food cost alone will never show it to you.
About this guide to menu labour cost
The short answer: labour cost and prime cost
Gross profit measures what a dish costs in ingredients. It says nothing about what it costs in time. Prime cost adds the two together: food cost plus labour cost, as a percentage of sales. A dish at 72 percent GP that takes twelve minutes of a chef's day can be less profitable than one at 66 percent that takes two, and nothing on a standard costing sheet will tell you.
Why food gross profit alone misleads you
Every costing model on this site and most others calculates margin on ingredients. That is correct, and it is incomplete, because ingredients are not the only thing a dish consumes.
Consider two mains priced identically.
| Dish A | Dish B | |
|---|---|---|
| Net selling price | £17.50 | £17.50 |
| Plate cost | £5.25 | £5.25 |
| Gross profit | 70% | 70% |
| Cash per plate | £12.25 | £12.25 |
| Prep and pass time | 2 minutes | 11 minutes |
| Labour at £15 an hour | £0.50 | £2.75 |
| Contribution after labour | £11.75 | £9.50 |
On a costing sheet these dishes are identical twins. In the kitchen one of them is costing you £2.25 more per cover. Across 150 covers a week that is £17,550 a year of difference between two dishes your model says perform the same.
The £15 an hour figure is illustrative. Use your own fully loaded rate, which means wages plus employer National Insurance plus holiday pay, not the headline hourly rate. Most operators underestimate this by a fifth or more.
What prime cost actually measures
Prime cost is food cost plus labour cost, expressed as a percentage of sales. It is the single most useful number in a hospitality P&L because it captures the two largest variable costs in one figure, and because the two trade off against each other constantly.
A kitchen can buy a lower food cost with more labour: butchering in house, making stocks, producing bread. It can buy lower labour with a higher food cost: pre-portioned proteins, bought-in sauces, prepped vegetables. Neither is right or wrong. What is wrong is optimising one while ignoring the other, which is what managing on food GP alone does by definition.
Why the published labour benchmarks disagree so badly
If you go looking for a labour cost benchmark you will find numbers that cannot all be true at once. It is worth knowing why before you measure yourself against any of them.
| Source | Figure given | Worth knowing |
|---|---|---|
| Shiftbase | 25 to 30%, full service nearer 30 to 35% | UK rota vendor |
| Loadedhub | Pubs 22 to 27%, casual 25 to 30%, fine dining 27 to 32% | Figures are Australia and New Zealand oriented |
| Toast UK | Around 15% | Far below every other source. Treat with caution |
The Toast figure is the obvious outlier and the most instructive one. A number that far below the rest of the field almost always means a different definition rather than a better operation.
The usual culprit is what counts as labour. Wages only, or wages plus employer National Insurance, holiday pay, pension and agency cover? Those two numbers can differ by twenty percent or more for the same business. Front of house included or kitchen only? Salaried management in or out? Before you compare your labour percentage to anything, write down your own definition, because the published figures rarely state theirs.
The same applies to productivity measures. The most commonly cited UK figure is from Fourth, which reported sales per labour hour across hospitality at £34.01, and £30.65 for restaurants alone. Fourth, 2017 That is from 2017. It predates several National Living Wage increases and the inflation of the last few years, so it is useful as a method and close to useless as a current target.
The honest position: there is no reliable current UK kitchen labour benchmark you can simply adopt. Your own trend over time is a far better instrument than anyone else's average.
How to work out what a dish costs in labour
You do not need a time and motion study. You need a defensible estimate, applied consistently.
- Split the time in two. Prep time is the share of a batch task attributable to one portion. Pass time is what the dish takes to fire, finish and plate during service. Both count, and they behave differently.
- Time the prep once, properly. Make the batch, note the minutes, divide by the number of portions it genuinely yields. A sauce that takes forty minutes and gives thirty portions is 1.3 minutes a plate.
- Time the pass during a real service, not on a quiet Tuesday. A dish that takes ninety seconds alone takes considerably longer when the section has eleven tickets up.
- Apply your fully loaded rate. Divide by 60 for a per minute cost, multiply by the dish's total minutes.
- Subtract it from cash margin. That is your contribution after labour, and it ranks dishes differently from gross profit almost every time.
Build the plate cost side of this with the recipe costing calculator, which already handles batch yields, then carry the per portion time alongside it in your specification.
Where menus quietly create labour
Most of the labour on a menu is committed at the point the dish is written, not at the point it is cooked. These are the decisions that cost the most and get examined the least.
- Garnishes with their own prep chain. A dish with four components each needing separate preparation is four prep tasks, four containers, four things to run out of mid service.
- Single use ingredients. An item prepped for one dish only carries all of its own labour and all of its own waste risk. An ingredient used across three dishes spreads both.
- Finishing that cannot be batched. Anything built to order at the pass is pure service time, and service time is the most expensive minute in the kitchen because it also constrains covers.
- Dishes only one person can cook. If a line depends on the head chef being on, it is not a menu item, it is a special with a permanent slot.
- Menu length itself. Every additional line adds prep, storage, stock and the risk of waste. A longer menu is a labour decision before it is anything else.
This is why a development brief should carry a prep time ceiling alongside the target plate cost. A dish that hits its cost and blows its time has not met the brief.
Cross utilisation: the cheapest labour saving available
Cross utilisation means using one prepped component across several dishes. It is the most reliable way to take labour out of a menu without taking anything off it.
A braised shin prepped once might appear in a main, a sharing plate and a lunch dish. One prep task, three selling lines, and the waste risk spread across three sources of demand rather than concentrated on one. The same thinking applies to trim: the salmon scotch egg built from fillet trim exists because the fish was already being prepped for another dish.
When you next look at a menu, count the distinct prepped components rather than the dishes. Two menus with fourteen dishes each can carry wildly different numbers, and the one with fewer components is cheaper to run, easier to train and less likely to drift from specification.
Common mistakes
- Loading labour into the plate cost. Do not. Gross profit is defined on food cost alone, and mixing labour in makes your GP incomparable with every benchmark and every other site in your group. Track it separately and look at both.
- Using the headline hourly rate. Employer National Insurance, holiday pay and pension are real costs of the hour. Excluding them understates dish labour by a fifth or more.
- Timing prep on a quiet day. The number you want is what the task takes under normal pressure, with interruptions, not under laboratory conditions.
- Comparing your labour percentage to a benchmark with a different definition. See the table above. Define yours first.
- Cutting labour by cutting hours. Taking an hour off the rota does not remove the prep; it moves it somewhere else or it does not get done to spec. Reducing the labour a menu requires is the durable version.
Where this fits with the rest of the numbers
Food gross profit tells you what a dish earns on ingredients. Pricing decides what it can charge. Food cost control explains why the theoretical figure and the actual one differ. Labour is the fourth number, and it is the one that decides whether a technically profitable menu is actually deliverable by the team you have.
A menu is a set of promises about what your kitchen will do every service. The menu engineering method ranks dishes on cash margin and sales mix. Adding time to that ranking is what turns an analysis into something a head chef will recognise as true.
Menu labour cost FAQs
What is prime cost in a restaurant?
Prime cost is food cost plus labour cost, expressed as a percentage of sales. It captures the two largest variable costs in one figure. It matters because the two trade off against each other: a kitchen can buy a lower food cost with more labour, or lower labour with a higher food cost, and managing either in isolation hides what is really happening.
Should labour be included in a dish's food cost?
No. Gross profit is defined on food cost alone, and including labour makes your GP incomparable with industry benchmarks and with other sites in your own group. Track prep and pass time separately against each dish, then look at contribution after labour alongside gross profit. Both numbers are useful; they are not the same number.
What is a good labour cost percentage for a UK restaurant?
Published figures range from about 22 to 35 percent of sales depending on format and source, and they disagree because they define labour differently. Some count wages only, others include employer National Insurance, holiday pay and pension. Define your own basis first, then track your trend rather than chasing someone else's average.
How do I calculate the labour cost of one dish?
Add prep time per portion to pass time per portion, then multiply by your fully loaded labour cost per minute. A dish taking six minutes in total at a fully loaded rate of £18 an hour costs £1.80 in labour. Subtract that from the dish's cash gross profit to get contribution after labour.
Does a shorter menu always cost less in labour?
Usually, but the number of distinct prepped components matters more than the number of dishes. A fourteen dish menu built from eight shared components is cheaper to run than a ten dish menu where every line has its own prep chain. Count components, not lines.
Is it better to buy in prepped ingredients or prep in house?
It depends which constraint is binding. Buying in raises food cost and lowers labour; prepping in house does the reverse. Work out the prime cost of each option rather than the food cost, and factor in consistency and the skill level actually available on the section. Neither answer is right in principle.
Keep reading. Build time matters most in lunch, breakfast and main courses. See also menu profitability by section.
How this page is sourced
The benchmark figures are attributed to the vendors who published them, with their orientation noted, because they do not agree and the disagreement is the point. The Fourth productivity figure is cited with its date because it is from 2017 and predates several wage increases.
The worked examples use illustrative figures and an illustrative hourly rate, not client data. No outcome is promised: what the labour in your menu is costing you depends on your menu, and establishing that is part of what a menu audit does.
A menu your kitchen cannot hold is not a profitable menu.
Send your current menu and last quarter's food GP. I will come back with the dishes I would look at first, on cost and on the time they take.