Blog · HoReCa

Food cost percentage: formula, calculation, and 2026 benchmarks

Stefan M. · marql · June 12, 2026 · Reading time: ~7 min

Food cost percentage tells you how much of your food sales is consumed by the ingredients used to produce those sales. If your restaurant sold EUR 95,000 of food and used EUR 30,000 of ingredients, your food cost percentage is 31.6%.

That number is simple. Using it correctly is not. A chain average of 31% can hide one location running at 26% and another running at 38%. One is disciplined; the other may be leaking margin through waste, over-portioning, supplier changes, or unregistered discounts.

In 2026, this matters more because ingredient pressure is uneven. The USDA ERS Food Price Outlook updated May 22, 2026 predicts food-away-from-home prices to rise 3.5% in 2026, with sharper pressure in categories like beef and veal, fresh vegetables, and nonalcoholic beverages. Your benchmark cannot be a static number in a spreadsheet. It has to be tracked against current recipes, invoices, and sales mix.


What food cost percentage means

Food cost percentage is cost of goods sold for food divided by food sales. It is not the same as total operating cost, and it is not the same as profit margin. It only answers one question: how efficiently are ingredients turning into revenue?

Food cost percentage = (food cost used / food sales) x 100

At restaurant level, "food cost used" usually comes from inventory movement: beginning inventory plus purchases, minus ending inventory. At menu-item level, it comes from recipe cost: the ingredient cost of one sold item divided by its menu price.

This distinction matters. A burger can have a clean recipe food cost of 29%, while the store that sells it runs at 36% because portions are inconsistent, waste is high, or invoices are posted to the wrong location.


Food cost percentage formula with example

Use the same period for all numbers. Weekly is better for operations because it catches problems while they are still small. Monthly is useful for accounting, but too slow if you manage several locations.

  • Beginning inventory: EUR 12,000
  • Purchases during period: EUR 28,000
  • Ending inventory: EUR 10,000
  • Food sales: EUR 95,000

Food cost used is EUR 12,000 + EUR 28,000 - EUR 10,000 = EUR 30,000. Divide EUR 30,000 by EUR 95,000 and multiply by 100. The result is 31.6%.

If your target is 30%, this looks like a 1.6 percentage point variance. On EUR 95,000 of food sales, that variance is roughly EUR 1,520 of margin that did not appear where expected.


Actual vs ideal food cost percentage

Actual food cost is what really happened. Ideal food cost is what should have happened if every recipe was portioned correctly, every ingredient price was current, and there was no waste, breakage, theft, or unrecorded consumption.

Ideal food cost percentage = (theoretical recipe cost of items sold / food sales) x 100

The gap between actual and ideal is the number operators should investigate. If a location has a theoretical food cost of 29% and actual food cost of 35%, the problem is probably operational, not pricing. If both theoretical and actual are 35%, the menu itself may be underpriced for the current supplier market.

This is why food cost control software should compare POS sales, inventory movement, supplier invoices, and recipe costs. POS revenue alone cannot tell you whether the kitchen protected margin.


2026 food cost percentage benchmarks

A useful benchmark is a decision range, not a universal rule. Restaurant365 notes that many restaurants aim around 28-32%, with quick-service concepts often lower and fine dining often higher. For 2026 planning, use the ranges below as a first screen, then tune them with your actual menu mix, vendor prices, and location economics.

  • Quick service and fast casual: 25-30%. Lower labor intensity and tighter portions usually support a lower target, unless premium ingredients dominate the menu.
  • Casual dining: 28-34%. This is the common operating band for many full-menu restaurants. Above it, check waste, portions, supplier prices, and discounting.
  • Full-service bistro or neighborhood restaurant: 30-35%. A slightly higher range is normal when recipes are more ingredient-led and the sales mix includes lower-margin mains.
  • Fine dining or protein-heavy concept: 34-40%. Premium beef, seafood, imported ingredients, and tasting-menu waste can push food cost higher. Margin must be protected through pricing and mix.
  • Cafe, bakery, dessert, and beverage-led formats: 18-28%. Ingredient cost may be lower, but packaging, waste, prep labor, and delivery commissions can still erode contribution margin.

Below range is not automatically good. It may mean pricing power, disciplined recipes, or a favorable sales mix. It can also mean portions are too small or product quality is being weakened. Above range is not automatically bad either, if the concept supports higher gross profit elsewhere. The question is whether the number is intentional, stable, and visible by location.


How multi-location operators should use the benchmark

Do not manage food cost from one chain-level average. A five-location restaurant group can report 31% total food cost while hiding a location at 39%. If that outlet does EUR 40,000 in monthly food sales, an eight point gap versus a 31% target is EUR 3,200 of monthly margin leakage from one store.

The right operating view shows food cost and gross margin per location, compared with the same day last week and the same week last month. It also separates sales mix from execution. If one store sells more low-margin items, the fix is menu mix and pricing. If it sells the same mix but uses more inventory, the fix is kitchen execution or stock control.

A correct daily sales report for restaurants should surface this signal every morning. You do not need a month-end P&L to know that yesterday's food cost moved outside the range.


Why food cost percentage goes up

When food cost rises, operators usually look at supplier prices first. That is sometimes correct, especially in 2026. But in a chain, the more useful question is: did every location move the same way?

  • Over-portioning. A 10% larger portion becomes a recurring hidden discount. It often appears first in one location or one shift.
  • Waste and spoilage. Prep mistakes, poor rotation, over-ordering, and unsold specials all increase actual food cost without increasing sales.
  • Supplier price drift. If beef, vegetables, dairy, or coffee move faster than menu prices, yesterday's profitable recipe can become today's margin leak.
  • Recipe mismatch. The recipe in the spreadsheet says 120 grams; the kitchen uses 150 grams. Theoretical food cost becomes fiction.
  • Discounts, comps, and voids. Food leaves inventory, but the sale does not land at full value in the POS. This is why POS and inventory data must be read together.

If all locations rise together, investigate vendor prices, recipes, and menu pricing. If only one or two locations rise, investigate execution, stock movement, voids, discounts, staff meals, and inventory counts.


What to track daily, weekly, and monthly

Daily tracking should be fast: sales, gross margin, average check, top products, and locations outside their normal range. This is enough to know where to look today.

Weekly tracking should compare actual vs ideal food cost by location and by category. This is where you identify waste, over-portioning, stock issues, and supplier variance before the month closes.

Monthly tracking should decide whether recipes or prices need to change. If the theoretical food cost of a popular item moved from 29% to 34%, the issue is not kitchen execution. The item economics changed.

marql connects to the systems you already use, including POS and accounting data, and turns them into one daily operating view across locations. See how the data flow works or view the dashboard structure. No POS replacement is required, and the first live view is positioned within 72 hours of the first connection.

Frequently asked questions

Food cost percentage

Ready when you are

Take control
of your stores.

Start with Starter or Growth self-serve, or book a 20-minute stack review for Scale and Enterprise. No POS replacement. No setup fee.

72h
Avg. onboarding
0
Replacement of stack
1
Cockpit. Whole chain.