Alberto de Palacio

Costs and profitability

Short menus and menu engineering: why the restaurants that make money have fewer dishes

A long menu fills your cold room and empties your till. How to use menu engineering to know which dishes to drop, which to push and which to put up.

A short menu earns more because every dish you drop is stock you don't buy, waste you don't throw away and kitchen time you don't lose. Menu engineering tells you which dishes to drop: you compare what each dish sells with what it leaves you every time you sell it. Four groups come out, and each one has a clear decision.

There's a temptation we've all had. A customer asks if you've got anything gluten-free, or some fish, or something “lighter”. And next month the menu has three more dishes on it.

And so, without noticing, you get to 45 dishes. And a cold room full of things that go out twice a week.

Why a long menu is costing you money

More stock. Every dish needs its ingredients. The more dishes, the more different things in the cold room, the more money tied up and the more risk of something going out of date.

More waste. The dishes that sell least are the ones that throw away most. The stock gets opened for one portion and the rest waits… until it's no good any more.

More work and more mistakes. More mise en place, more spec sheets to remember, more chance of each dish coming out differently depending on who's on the pass.

Customers who take longer to decide. An endless menu doesn't give freedom, it gives stress. The customer takes longer, asks more questions and often ends up ordering the usual.

A blurred identity. If you do a bit of everything, nobody knows why they should come to your place. And in 2026, with customers choosing more carefully where to spend their money, having a clear concept is one of the things that weighs most.

What menu engineering is

It's a method for analysing the menu with numbers, not with sentiment. Two American professors, Kasavana and Smith, developed it in the eighties, and it's still used because it works.

It compares two figures for each dish:

  1. Popularity: how many portions you sell compared with the rest.
  2. Margin: how much each portion leaves you (price without VAT minus the cost from the recipe costing).

And it puts each dish into one of these four groups:

High margin Low margin
Sells a lot Star Plowhorse
Sells little Puzzle Dog

How to do it with your menu, step by step

Step 1. Gather a month's data

For each dish in one section (starters, mains, desserts… each section is analysed separately):

Step 2. Work out each dish's margin

Margin per portion = Price without VAT − Cost of the portion

Notice that we're not looking at food cost as a percentage, but in euros. A dish with a 40% food cost can leave you more euros than one at 25%. And what you pay at the end of the month is euros.

Step 3. Separate the popular dishes from the rest

The method's rule: a dish is popular if it sells more than 70% of what it would get if every dish sold the same.

In practice: divide 100 by the number of dishes and multiply by 0.7. With 6 dishes: 100 ÷ 6 × 0.7 = 11.7% of the section's sales. Anything above that is popular.

Step 4. Separate high margin from low margin

Work out the section's average margin: add up the total margin of all the dishes and divide by the total portions. Anything above it is high margin. Anything below, low margin.

Step 5. Classify and decide

Now every dish is in its box. Time to act.

An example with numbers

A mains section with 6 dishes, in a typical month (an illustrative example, prices without VAT):

Dish Portions Price Cost Margin/portion Total margin
Homemade croquettes 400 €9.00 €2.20 €6.80 €2720
House burger 350 €13.00 €5.50 €7.50 €2625
Señorito rice 180 €16.00 €5.60 €10.40 €1872
Grilled octopus 90 €19.00 €7.20 €11.80 €1062
Seasonal salad 70 €10.00 €3.10 €6.90 €483
Lasagne 60 €12.00 €4.60 €7.40 €444
Total 1150 €9206

Result:

Dish Popular? High margin? Group
Señorito rice Yes (180) Yes (€10.40) Star
Homemade croquettes Yes (400) No (€6.80) Plowhorse
House burger Yes (350) No (€7.50) Plowhorse
Grilled octopus No (90) Yes (€11.80) Puzzle
Seasonal salad No (70) No (€6.90) Dog
Lasagne No (60) No (€7.40) Dog

Look at the croquettes. They're the dish that sells most. Everyone would say they're the best thing on the menu. And they're the one that leaves least per portion.

What to do with each group

Stars: protect them

They sell and they leave money. Don't touch them, except to look after them.

Plowhorses: raise their margin without anyone noticing

They sell a lot but leave little. They're the ones that can give you the most money if you adjust them a little, because every cent is multiplied by a lot of portions.

In the example: if the croquettes go up by 50 cents, that's €200 more a month without selling a single one more.

Puzzles: give them a chance

They leave a good margin, but they don't sell. Often the problem isn't the dish, it's how it's sold.

If after a couple of months it still isn't selling, it moves to the group below.

Dogs: off the menu

They neither sell nor leave money. All they do is take up cold room, time and space on the menu.

In the example, the salad and the lasagne come off the menu. Two dishes fewer, two recipes fewer to prep, and all their stock out of the cold room.

How many dishes should my menu have?

There's no magic number; it depends on the concept and the kitchen you have. My reference: a few options per section, which you can make perfectly every day. If your kitchen can't turn out a dish just as well on a Saturday at ten as on a Tuesday at two, that dish is surplus.

And a good sign: if someone reading your menu knows in ten seconds what kind of restaurant you are, you're on the right track.

And what about average spend?

A short, well-worked menu also raises average spend, because:

How often to repeat the analysis

Do you know which of your dishes are stars and which are dogs? Start with the cost of each portion using the recipe costing template. And if you'd like us to analyse your menu together, let's talk.

Tell me about your business

Frequently asked questions

What is menu engineering?

It's a method for analysing the menu by comparing each dish's popularity with the margin it leaves. It classifies dishes as stars, plowhorses, puzzles and dogs, and helps you decide what to keep, what to adjust and what to drop.

Do I need recipe costings to do menu engineering?

Yes. Without knowing what each portion costs you, you can't work out the margin, and without margin the analysis is useless.

Is it better to look at food cost or margin in euros?

Margin in euros. The percentage tells you whether a dish is well costed, but what you pay at the end of the month is euros. An expensive dish with a high food cost can leave you more money than a cheap one with a low food cost.

How often do I have to change the menu?

There's no need to change the whole thing often. Check the numbers every quarter, drop the dogs, work on the puzzles and adjust the plowhorses. Two menu changes a year, one for summer and one for winter, are usually enough.

Host Brain

Want to automate your restaurant?

A team of AI agents watches your costs every day, and all you do is approve or dismiss.

Meet Host Brain