Restaurant Food Cost Percentage: What It Is and How to Calculate It

Every restaurant operator knows their food cost percentage. Or they think they do. In my experience working with restaurant financials, most owners are tracking a number that's either stale, calculated wrong, or missing the point entirely. Food cost percentage is useful, but only if you understand what it's actually telling you.

What Is Food Cost Percentage?

Food cost percentage is the ratio of what you spend on food ingredients to what you make selling those ingredients as menu items. It's expressed as a percentage of food sales. If you spend $8,500 on food in a week and sell $30,000 in food, your food cost percentage is 28.3%.

That number tells you how much of every dollar from food sales goes back out the door to buy ingredients. The rest has to cover labor, rent, utilities, insurance, and whatever profit is left.

How to Calculate It

The formula:

Here's a real example. Say you start the week with $5,000 in inventory, buy $8,000 more, and end with $4,500. Your cost of goods used is $8,500. If food sales were $30,000:

$5,000 + $8,000 − $4,500 = $8,500
$8,500 ÷ $30,000 = 28.3%

You can also calculate food cost per dish. If a burger costs you $3.50 in ingredients and you sell it for $12, the food cost percentage on that item is 29.2%. This is called theoretical food cost — what the recipe says it should cost.

What's a Good Food Cost Percentage?

The industry benchmark is 28% to 35% of food sales. But that range varies a lot by segment:

• Quick service (QSR): 25% to 30%. Standardized menus, high volume, lower ingredient costs per unit.
• Fast casual: 27% to 32%. Better ingredients and more customization than QSR.
• Full-service casual dining: 28% to 32%. Wider range depending on menu complexity.
• Fine dining: 30% to 40% or higher. Premium ingredients, lower volume, more waste tolerance.

These ranges come from National Restaurant Association data and operator surveys. They're benchmarks, not targets. A 32% food cost at a steakhouse is different from 32% at a sandwich shop. The context matters.

Theoretical vs. Actual: Where Money Leaks

This is the part most operators miss. There are two food cost numbers: theoretical and actual.

Theoretical food cost is what your recipes say each dish should cost, multiplied by how many you sold. It assumes zero waste, perfect portions, no spoilage, no theft.

Actual food cost is what your inventory numbers show you actually spent. It's calculated from the formula above.

The difference between the two is called variance. A variance of 1% to 5% is normal. Nobody is perfect. But if your theoretical food cost is 29% and your actual is 36%, you have a 7% gap. That gap is real money leaving your business through over-portioning, waste, spoilage, comps, or theft. Most operators never calculate theoretical food cost, so they don't know the gap exists.

Prime Cost: The Number That Actually Matters

Here's the thing that took me a while to learn: food cost percentage alone doesn't tell you if your restaurant is profitable. You need prime cost.

Prime cost = food cost + labor cost

The benchmark for prime cost is 55% to 65% of total sales. Most operators target 60% or below.

Consider two restaurants:

• Restaurant A: 32% food cost, 35% labor cost. Prime cost = 67%. This restaurant is probably losing money.
• Restaurant B: 35% food cost, 25% labor cost. Prime cost = 60%. This restaurant is probably profitable.

Restaurant B has a higher food cost percentage but is in better shape. If you only look at food cost, you'd think Restaurant A was doing better. It's not.

What Operators Get Wrong

Using One Target for the Whole Menu

Different menu items have different food cost percentages by design. A steak entree might run 40% food cost. A pasta dish might be 18%. That's not a problem — it's menu engineering. The goal isn't to hit one percentage on every plate. The goal is to balance high-cost, high-margin items across the menu so the blended food cost lands in range.

Not Raising Prices When Costs Rise

Wholesale food prices are elevated. According to the National Restaurant Association, the overall Producer Price Index for food was only up 0.1% year-over-year as of May 2026. But that aggregate number masks massive swings in specific commodities. Beef and veal prices were up 15.9%. Fresh vegetables were up 123.2%. Coffee was up 8.8%. If your menu is heavy on those categories, your actual cost increase is far worse than the headline number suggests.

Operators who absorb those increases without adjusting menu prices are quietly eroding their margins. Small, regular price adjustments of 1% to 3% are better than waiting and doing a 10% jump that sends customers to the competition.

Confusing Food Cost % with Food Cost Dollars

A higher food cost percentage can still mean more profit. A $40 steak at 38% food cost gives you $24.80 in gross profit. A $15 burger at 28% food cost gives you $10.80. The steak has a worse food cost percentage but puts more money in your pocket per sale.

Gross profit dollars matter more than the ratio. The ratio is a diagnostic tool, not the finish line.

Ignoring Inventory Shrinkage

If you're not doing regular physical inventory counts and reconciling them against your POS data, you don't know your actual food cost. You're guessing. Shrinkage of 2% to 4% of food cost is common, and most of it is preventable.

How to Get It Right

Start with these three things:

1. Do weekly inventory counts. Weekly counts catch problems before they compound into monthly losses.
2. Calculate theoretical food cost per dish. Cost out every recipe. Compare the blended theoretical to your actual. Find the gap.
3. Track prime cost, not just food cost. If food + labor is above 65% of sales, you have a profitability problem regardless of what either number looks like alone.

Most restaurants that struggle aren't failing because their food cost is 33% instead of 30%. They're failing because they don't know their actual food cost, don't know their theoretical, and don't track prime cost. The numbers are all available. Most operators just aren't running them.