What is food shrinkage: traditional method vs Masterestaurant method

What is shrinkage (merma) in a restaurant?

Shrinkage is the quantity of an ingredient the restaurant pays for that never reaches a charged portion, measured by weight or volume between the receiving scale and the plate that goes out.

Put that way it sounds like accounting, but it is pure kitchen: the bone left on the cutting board, the water the chicken releases in the oven, the cilantro that turns black in the walk-in before Tuesday and the extra spoonful a cook plates when the rush hits. All of it was bought at the gross price per kilo and is sold at the net price per portion, and the gap between those two numbers is the shrinkage.

That is why the useful definition for an owner is not the dictionary one but the one on the recipe card, because there shrinkage stops being a chef's complaint and becomes a cost that someone signs with their name.

The four families of shrinkage and who answers for each

Shrinkage has no single owner, and treating it as one number is the reason nobody fixes it. At Masterestaurant we split it into four families. Trim loss starts at purchasing: it depends on the specification you demand from the supplier and on whether the fish arrives scaled or whole. A badly calibrated oven, on the other hand, steals weight from a pork leg without anyone noticing, which is why cooking loss belongs to the recipe and the equipment.

Whatever spoils in the walk-in is the storekeeper's business and depends on the discipline of first in, first out. And service loss lives on the line, in portions plated by eye and in the extra batch made just in case: ReFED attributes 11.9% of the surplus from U.S. restaurants and foodservice in 2024, about 1.49 million tons, to overproduction.

How to calculate shrinkage: an avocado example?

Shrinkage is calculated by weighing the ingredient gross, cleaning it, weighing it again net and dividing net by gross: that ratio is the yield factor, and it is what corrects the cost per kilo.

Take a 10-kilo case of avocado bought at the per-kilo price on your invoice, and multiply it by the full weight of the case to see what you actually paid. Without skin and pit you are left with only part of the case as usable flesh, so the yield is lower than the weight you bought and each kilo of flesh costs you well above the invoice price per kilo.

A guacamole portion with 120 grams of flesh therefore costs 11.08 pesos, not the 7.20 you get if you cost it at the invoice price. That is 3.88 pesos hidden in every portion, and at 40 portions a day the example adds up to 155.20 pesos daily that no sales report will ever show you. Repeat the test in different weeks before you lock the factor into the recipe card.

What shrinkage is NOT: interpretation errors that cost money?

The most expensive error is believing the goal is zero shrinkage, when part of it is natural and gets budgeted: nobody charges the guest for the avocado skin, yet it has to be in the price.

Another frequent stumble is expressing it as a percentage of sales, which mixes a yield problem with a pricing problem and leaves the figure without an owner, when shrinkage is measured against the weight of the ingredient. Nor is it the same as the accounting gap in inventory, which may come from a badly entered invoice.

And the bar is almost always forgotten, even though it shrinks too: in Spain, 54.5% of food waste outside the home is food and the rest is drinks and snacks (Ecoticias, with data from the Ministry of Agriculture, Fisheries and Food, 2024). The half-squeezed lime and the open bottle that oxidizes count just like the beef trim.

Why is measuring shrinkage at month-end too late?

Because by month-end the shrinkage has already been sold at the wrong cost in hundreds of plates, and inventory only tells you how much you lost, never in which ingredient or on which shift.

Think about what would happen if the real yield of your chicken breast were 70% and your recipe card said 85%: every portion would leave the kitchen cheaper on paper than on the scale, the menu price would be set on that false cost, the dish that looks like a star would be the one eating the most margin, and when inventory arrived you would see a hole nobody can explain because thirty days have already passed.

The scale of the problem is not anecdotal. ReFED counted more than 8.2 million tons of unsold or uneaten food in full-service and limited-service U.S. restaurants in 2023. That is why Diego F. Parra insists on measuring BEFORE, with the yield test, rather than blaming the kitchen afterwards.

Shrinkage at regional and global scale

A kitchen's shrinkage is the part you control of a much larger problem, and it pays to give it a size so you stop treating it as a detail. In Latin America and the Caribbean, 11.6% of the food produced is lost, according to the FAO, and the same regional office puts the economic cost of those losses at 150 billion dollars every year.

At global scale, UNEP's Food Waste Index Report 2024 documented an enormous amount of food wasted in 2022, a large share of it coming from consumption in homes and food service. None of those figures is your shrinkage, and it would be a mistake to use them as a target for your restaurant.

They help you understand that the ingredient rotting in your walk-in is part of a chain where every lost kilo already carried water, transport and labor. Your own number only comes from the scale, ingredient by ingredient, and nobody can hand it to you.

Should you buy the product already cleaned to cut shrinkage?

Sometimes yes and sometimes no: the right decision comes from comparing the NET cost per kilo of both options, never the invoice price or the shrinkage on its own.

That is the paradox that confuses many owners, because the clean fillet has almost no shrinkage and can still be the more expensive purchase, while the whole product sometimes turns out cheaper per portion. A single division reconciles the two ideas.

If the whole product, at its real yield, comes in below the clean one, shrinkage is good business and leaves you bones for a stock; if it comes in above, pay the supplier to clean it and free up hours in your kitchen. Masterestaurant runs this math before every supplier change, and it is what keeps the 32% food cost ceiling per plate an honest maximum.

Small decisions like this, repeated across thousands of kitchens, sit behind the 28.03 million kilos or liters that Spanish hospitality wasted in 2024 (Ministry of Agriculture, Fisheries and Food).

The differences that move your margin

The traditional method measures shrinkage after the fact: at month-end it finds a hole nobody can explain, because hundreds of plates have already been sold at the wrong cost. At Masterestaurant we measure it BEFORE, with yield tests per item, and the resulting factor goes straight into the recipe card, so the price is born with shrinkage built in. Traditional costing treats all shrinkage as one number.

We split it into four families with different owners: trim loss belongs to purchasing and specs, cooking loss to the recipe and equipment, storage loss to the storeroom and its FIFO discipline, and portioning loss to the line cook. One blended percentage guarantees that nobody takes charge.

The differences that move your margin — in practice

The loss that hurts most often is not on the cutting board but in overproduction. ReFED attributes 11.9% of US restaurant and foodservice surplus in 2024 to it, the batch of rice or beans made just in case that ends up in the bin on Sunday. Diego F. Parra puts it at the center, with production driven by the sales forecast instead of habit.

Here I was wrong for years: I also believed a flat buffer added to every recipe solved shrinkage. It works on a spreadsheet and fails in the kitchen, because avocado and tenderloin do not lose the same, and a uniform buffer hides exactly the items that cost you the most while inflating the ones that barely shrink.