How to know if your restaurant is profitable when you keep the numbers in your head

They have asked you a thousand times and you answer the same way, tapping your temple.

It is not bravado. You know what a case of tomatoes costs, which days people come in, and within about a hundred euros what will be in the till on a Friday in October.

You have done it this way for years and the business is still open. That is data too.

The question is not whether your head works. It is which part of the business fits inside it.

The head tracks what repeats and what hurts: rent, payroll, the price of the four products you buy every week.

What it does not track is the slow drift.

One supplier raises meat 4% in March, another wine 3% in June, another the garnish 5% in September. None of the three makes you look up from the delivery note.

Together they make the dish you sell most about 12% more expensive, and your menu still carries January’s price.

You do not see that in the till. You see it three months later, when the accountant closes the quarter, the number comes in worse than you expected and there is no way to know which of the forty dishes ate it.

Here is the figure, and it comes with the fine print first.

In the study “Digitalisation in the tourist HORECA channel in 2026”, which Hostelería de España and SEGITTUR published in June 2026, Paco Cruz, hospitality consultant and director of The Food Manager, estimates from years of training operators that around 90% have no profit and loss statement.

That is not a survey. It is one professional’s estimate after years of training in the sector, recorded inside an institutional report. There is no sample or margin of error behind that number, and anyone selling it as if it came from the national statistics office is selling it badly.

What you do have is a second figure, measured with a different stick, pointing the same way: in the traditional operator profile that study draws, only 50% review monthly whether the business is profitable.

A profit and loss statement is the list of what comes in and what goes out in a month, ordered, excluding VAT and counting product consumed instead of product bought. That is all it is.

Do not confuse this 90% with another different figure Tipi also uses: the 90% without an annual budget that appears in why you cannot fully delegate control of the numbers. Same figure, different meaning: there Adrià is talking about budgeting; here it is a field estimate about profit and loss statements.

Spain has more than 300,000 hospitality establishments, according to the Spanish Hospitality Yearbook.

Apply that 90% and you get about 270,000 venues run without knowing whether they make money. Our own sum on someone else’s estimate: useful for scale, not to cite as a census.

Put it beside a different figure. In Madrid, according to Fernando Múgica’s closure count that El Español published in May 2025, 63% of venues that shut had not reached five years.

Those are two separate things and nobody has shown that one causes the other. But it is the ground you work on.

If you tried something and abandoned it, you are not the exception. You are the common case.

73% of operators say digitalisation has not saved them even half the time they spend on management, according to TheFork data gathered in the study.

And 30% name lack of digital training as the main obstacle.

The study itself closes with a line from one of the operators interviewed: “it is not lack of interest, it is lack of time and support”.

Which is another way of saying they did not sell you a method. They sold you a screen.

A worked example with round numbers. Invented menu to explain the mechanism, not any venue’s recipe costing.

Cheek. You put it at €16.50 in January, with a cost of €5.20. Food cost 31.5%, which is fine.

By September the meat has risen €0.90, the wine in the sauce €0.20 and the garnish €0.15.

The dish now costs €6.45 and the menu still says €16.50. Real food cost is 39%.

That is €1.25 less per plate. If you sell 25 a week, that dish leaves €1,625 less a year than in January. One dish. Out of forty. That is the money that shows up nowhere until someone goes looking for it.

The Hostelería de España and SEGITTUR study classifies operators into three maturity levels. Translated into second person:

The distance from the first level to the second is not buying software: it is costing ten dishes once.

If what you lack is the cadence (weekly, monthly, quarterly), not the first thermometer, we unpack that in how often you should review restaurant numbers.

Once you have that result, the next question is how much of that money you can actually take out: we unpack it in full in the article on real restaurant profit.

It is qualitative research, not a representative survey. 49 interviews with key actors, 6 focus groups, 18 day-to-day shadowings and 97 channel agents. Solid for understanding why things happen, not enough to publish national percentages.

The 90% is a consultant’s field estimate, not a measurement with a sample.

The closure figure is Madrid and a private count, not an official register or a national series. And it does not say venues close because they fail to control costs: it says how many closed and how long they lasted.

And the cheek example is ours. Your menu has different numbers.

Tipi is listed among the technology companies that collaborated on the study. That does not change the fine print on the data; it does explain why we know it first-hand.

You set the menu price. It is the only part of this whole story that does not depend on a supplier.