Food Cost Percentage: The One Number Most Restaurants Are Calculating Wrong

Ask ten restaurant owners what their food cost percentage is and eight of them will give you a number. Ask them when they last calculated it, and at least half will admit it was a while ago. Ask them how they calculated it, and that's where it falls apart.

There's a version of food cost that's technically a number but operationally useless. It's the one most bookkeepers run: take your food purchases for the month, divide by your food sales for the month, multiply by 100. Done. It's 32%. Good job.

That number is almost always wrong.

Not because the math is wrong. Because the inputs are wrong. And because the food cost you can actually manage against isn't a monthly number. It's a weekly number, and sometimes a daily one.

Let me walk through why.

The formula everyone knows, and why it lies

Standard food cost is: (Beginning inventory + Purchases – Ending inventory) ÷ Food sales.

If you skip the inventory part (which most small operators do), you're calculating purchases-to-sales, not actual cost of goods sold. These are different numbers. Sometimes by a lot.

If you bought $18,000 in food in March but ate into $3,000 of inventory you already had on hand, your real food cost was $21,000. Your food cost percentage is higher than the simple formula suggests. You're less profitable than you think.

Flip it around. If you bought $18,000 but added $2,500 of inventory to the walk-in that carried into April, your real food cost was $15,500. You're more profitable than the number says. But you're about to panic when April's numbers look worse than March's, even though nothing actually changed.

This is the whole reason restaurants feel like they can't tell whether they're making money. The number moves, and they don't know why. Usually: it's inventory swing, not operational change.

Why weekly beats monthly

Monthly food cost is a report card. Weekly food cost is a steering wheel.

If you calculate food cost once a month, by the time you see a problem, you've already lost four weeks of margin. You can't go back and unorder the ribeye you bought at the wrong price. You can't un-trim the portion you cut too generously.

Weekly food cost, run every Monday or Tuesday for the prior week, gives you a chance to react. You see the 38% number on Tuesday morning. You pull the invoices. You see that protein cost spiked 11%. You call the rep. You swap the special. By the end of the week, you're back at 31%.

Operators who run weekly food cost tend to sit 3–5% below operators who run monthly. That's not magic. It's just that they catch problems in week one instead of week four.

The inventory you're not counting

Most small restaurants count the big stuff (proteins, produce, sometimes dry goods) and ignore everything else. That's fine, up to a point. The Pareto on this is real: 70% of your food cost is usually 20% of your SKUs.

But there are three categories that get ignored and shouldn't:

Opened bulk items. That 50-pound bag of flour that's half gone? It's inventory. So is the mostly-empty tub of butter. Partial inventory counts feel silly, and eyeballing is fine for most items. But if you're consistently underestimating partials, you're overestimating ending inventory, which understates food cost.

The walk-in's hidden corners. Every walk-in has them. Stuff shoved behind other stuff. Portioned proteins in vac bags that are technically labeled but that nobody's touched in a month. These count. Even if they shouldn't be there anymore. Until they're thrown out, they're on the books.

The back bar and the line. If you're counting for food cost, count the prep line and the back bar too. That bin of cut onions, the speed rack of plated-up butter, the rolled silverware with lemons wedged in. Some of that is inventory, some is staging, and it matters for how you interpret the number.

What "food cost" actually includes

Here's where operators disagree, and it matters because comparisons get muddy.

Food cost, strictly, is cost of goods sold for food items. But in practice, most operators roll a few other things in, and a few things out, in ways that aren't standardized.

What's usually in: raw ingredients, portioned proteins, produce, dairy, dry goods, non-alcoholic beverages if they're included in food sales.

What's sometimes in, sometimes not: oil for the fryer (it's a consumable, but it's a real cost), disposables like to-go containers (arguably a packaging line item, not food), spices and staples, employee meals.

What's almost never in: alcohol (that's beverage cost, and it runs separately), labor, overhead.

None of this matters if you're internally consistent. What kills you is changing the methodology mid-year. If in January you counted disposables as food cost and in June you moved them to packaging, your numbers aren't comparable. You'll think you improved by 2% when you really just moved costs from one column to another.

Pick a definition. Write it down. Stick to it.

The three biggest leaks

If your food cost is higher than you'd like, it's almost always coming from one of three places. In rough order of how much they matter:

1. Portioning. This is by far the biggest one. A 7-ounce pour of a 6-ounce recipe, done twice a night across two line cooks, adds up to a couple thousand dollars a year on that one item. Portioning drift is almost never intentional. It's habit. Tired cooks scoop heavier. New cooks don't know what the target looks like. Managers stop weighing. Eventually the recipe is whatever the line is doing.

A portion scale on the line, and a weekly spot check by a manager who actually weighs plates, fixes most of this. It's dull work. It's the most effective thing you can do. Our restaurant operator SOP pack includes a line-check SOP built around this.

2. Waste. Trim waste, prep waste, over-prep, spoilage. Some of this is unavoidable. A lot of it is tracking failure. If you don't log waste daily, you have no idea how much you're throwing away, and without that, you can't improve it.

Waste logs are annoying, I know. But they don't need to be fancy. A clipboard by the trash. Three columns: item, amount, reason. That's it. At the end of the week, you'll see patterns that'll surprise you.

3. Over-ordering. Restaurants get into a rhythm of ordering what they ordered last week plus a little extra for safety. That extra adds up. If your walk-in is consistently 80% full on delivery day, you're over-ordering. Your par levels are probably wrong. Fix the pars, not the count.

The number you should actually be watching

Food cost percentage is a ratio. It's affected by both the numerator (cost) and the denominator (sales). That's a problem, because the two move independently, and a "good" food cost number can hide a real issue.

If sales are up 20% because you ran a big promo at a lower margin, your food cost percentage might spike even if nothing changed in the kitchen. Conversely, if sales tank, your food cost percentage can look great while you're losing money on volume.

The number you actually want to watch is food cost per guest, or food cost per cover. Take your food cost for the week, divide by your number of covers. That's a cleaner signal. It strips out pricing effects. It tells you, on average, how much food each guest is eating, and it moves for real reasons: menu mix shifts, portioning changes, ingredient prices.

Track it alongside food cost percentage. When they move in different directions, that's where the interesting story is.

What to do Monday morning

If you're running a small restaurant and you want to get your food cost under control, here's what I'd do in the first two weeks:

Do a full inventory count on Sunday night. Yes, all of it. Yes, including partials. Pay someone if you have to.

Calculate your actual food cost for the prior week using real beginning and ending inventory. Not purchases-to-sales. Real COGS.

Write down the number. Post it somewhere you and your managers will see it.

Repeat next Sunday night. Compare the two weeks. Whatever changed between them is your starting point.

A lot of operators look at this process and decide it's too much work to do weekly. Fine. Go monthly if you must. But be honest with yourself about what you're trading off. You're trading off the ability to react in time to matter.

The operators who are making money in this industry are not smarter than the ones who aren't. They're just counting more carefully, more often, with cleaner inputs.

---

Our restaurant profit & operations manager workbook includes a weekly food cost tracker, an inventory count sheet built for actual prep kitchens, and a par level worksheet. It's built by people who've done this from the line side, not from a consulting deck.