Your menu mix

This is the topic that pays for the course. Every menu contains dishes that sell brilliantly and leave almost nothing, and dishes that leave a fortune and nobody orders — and because those two facts live in different reports, most owners never put them next to each other. Doing so takes an afternoon, needs no software, and routinely finds several hundred pesos a month that were already on the menu.

✖️ Two numbers per dish, multiplied

For each dish you need how many you sold in a period, and what it leaves you after its ingredients — the price minus the food cost, which is the contribution. Not the percentage: the pesos. That distinction matters more than anything else in this topic, because you bank pesos, not percentages, and a dish with a beautiful cost percentage can leave less money in the till than an ugly one.

Here is a month of four dishes from the example restaurant.

Read the last column and the whole point of the exercise appears. The burger is the best-seller by a distance — a hundred plates more than the pasta — and it contributes seven hundred pesos less. Nobody standing on the line would ever notice that, because the line experiences the burger as the thing that never stops. The report is the only place that fact exists.

🗺️ Four kinds of dish

To sort the menu you need two dividing lines: the average number sold, which here is about 278 dishes, and the average contribution, which is $12,760 divided by 1,110 dishes, or about $11.50. Every dish falls on one side or the other of each, which gives four groups — and the four have completely different treatments.

The workhorse is where the money is, and it is the one owners handle worst. Removing your best-seller because its margin is thin is a catastrophic move — people come for it. But because it sells five hundred times a month, an extra peso of contribution on it is five hundred pesos, which is more than most restaurants save in a month of fretting about suppliers. A slightly smaller portion of an expensive component, a cheaper garnish that does the same job, or a one-peso price move all get you there without touching what people love about it.

The puzzle is the other big opportunity, because the work is selling rather than cooking. The fish leaves eighteen pesos, the most on the menu, and almost nobody orders it — so the entire question is whether it is invisible, badly described, badly positioned, or genuinely just not wanted. Moving fifty burger orders to pasta orders in a month is worth two hundred pesos; getting the fish ordered more often is worth more per plate than anything else you can do.

💰 Percentages lie, pesos do not

Most kitchens judge a dish by its food cost percentage, and it is genuinely useful for buying and for spotting a recipe that has drifted. As a way of deciding what to sell, though, it points in the wrong direction often enough to be dangerous, because a percentage is a ratio and your bank account is not.

Look at two dishes from the table. The pasta costs five pesos on an eighteen-peso price, which is a food cost of about 28% — the best-looking number on the menu. The fish costs fourteen on thirty-two, nearly 44%, which in most kitchens would be flagged as a problem. Yet the fish leaves eighteen pesos and the pasta leaves thirteen. Sell one fish instead of one pasta and you are five pesos better off, while your food cost percentage gets worse.

This is why the whole topic is built on pesos of contribution rather than percentages. It also explains a pattern that confuses owners: a month can show a rising food cost percentage and a rising result at the same time, simply because customers moved towards the expensive dishes that leave more money. The percentage belongs in the cost control course, where the question is what things cost you. Here the question is what they leave you, and that is always a peso figure.

🍽️ Acting on it without wrecking the menu

Once the four groups are on paper the temptation is to redesign everything at once, which is how restaurants lose the thing that made them work. Move slowly, one or two dishes at a time, and give each change a month before judging it.

• Cut dogs, but check them first. A dish with poor numbers can still earn its place: it might be the only option for a dietary need, the thing one big regular table always orders, or a shared ingredient that keeps another dish's cost down.
• Shrink the menu before you expand it. Fewer dishes means less stock, less waste, faster service and better execution — and the sales rarely disappear, they redistribute to what is left.
• Position matters more than you think. Where a dish sits on the page and whether it has a description moves its sales meaningfully, and it is free.
• Tell the floor which dishes matter. Servers recommend what they know and have tasted. A star and a puzzle named at pre-shift will outsell a menu redesign.
• Re-run it after any supplier price change. Contributions move when costs move, and a star can become a workhorse without anything visible happening.
• Do it by section. Starters compete with starters. Comparing a dessert against a main course tells you nothing useful.

Two boundaries worth respecting while you do this. How to work out a dish's cost properly, and how to bring that cost down, is the cost control course — this one takes the cost as given. And how to set the price itself is a discipline of its own; here a price only changes as a small, deliberate adjustment to fix a workhorse, not as a strategy. What this topic owns is the crossing of the two numbers, and that crossing is where the money you already have is hiding.