Restaurant Food Cost Weekly, Not Yearly: The Prime Cost Habit

Restaurant food cost percentage is a weekly number or it’s a story. Seen once a year on the tax return, a 32% food cost is an average that hides the two months it ran 39% because a vendor raised prices and nobody caught it. Seen every week, from invoices and a count, the spike shows up the week it happens, while there’s still time to fix the menu, the portion, or the vendor. Add labor and you have prime cost, and prime cost under 60% of sales is the number a restaurant lives or dies on.

We run an accounting firm, and we just closed four full years of restaurant books in under ten hours of staff time before the September 15 deadline. What those books showed, once they were monthly, is the reason for this article.

What should food cost percentage be for a restaurant?

Most concepts run food and beverage cost somewhere between 28% and 35% of sales. A pizza place sits at the low end. A steakhouse sits at the high end. A bar’s beverage cost is lower than its food cost. The target depends on the concept, and the target matters less than the trend.

The number is purchases plus beginning inventory minus ending inventory, divided by sales, for the same period. That “same period” is where yearly numbers fail. Food bought in the last week of September for October’s business lands in September’s purchases and October’s sales. Over a year it washes out. Over a week or a month, it doesn’t, which is why the count matters.

How often should a restaurant calculate food cost?

Weekly. Here’s the routine, and it’s shorter than it sounds.

Purchases from invoices. Every food and beverage invoice for the week, by vendor, entered in the books with the vendor on every line. Not “Sysco $4,212.” Sysco, with the invoice split at least between food, beverage, and non-food supplies. The split takes a minute at entry and is impossible to reconstruct later.

A count. Sunday night or Monday morning, the same time every week, the walk-in, the dry storage, the bar. It doesn’t need to be to the ounce. It needs to be consistent, because the trend is what you’re reading.

Sales from the POS. Net sales for the week, food and beverage separately if the POS can split them.

Cost of goods = beginning count + purchases – ending count. Divide by sales. Write it on the whiteboard in the office. Seven days later, do it again.

A restaurant that does this sees a 4-point jump the week it happens and can trace it in an afternoon: the produce vendor’s new price list, the new cook’s portions, the special that was priced wrong, the walk-in that lost a case of chicken. A restaurant that does it yearly sees a 1-point difference on the tax return and a shrug.

Labor

Labor is the other half, and it’s the half that moves faster. Weekly, from the schedule and payroll:

• Hourly wages for the week, front and back of house separately if you can.
• Salaried management, one-fifty-second of the annual figure.
• Employer payroll taxes and workers’ comp on all of it.
• Tips handled as tips. Tips are the employee’s money passing through you, not labor cost. If tip payouts are running through the books as wages, or as checks to individuals booked as “contract labor,” your labor cost is wrong and so is your payroll filing. The four-restaurant cleanup this month involved close to 700 handwritten checks at one location, and separating tip payouts from vendor payments was most of the work.

Labor as a percent of sales, weekly, tells you whether the schedule matched the business. A slow Tuesday with a full crew shows up as a bad week, not a bad quarter.

What is a good prime cost for a restaurant?

Prime cost is food and beverage cost plus labor cost, as a percent of sales. Under about 60% works. Between 60% and 65% is survivable if occupancy costs are low. Over 65% is structural, and volume won’t save it, because every extra dollar of sales brings 65 cents of cost with it before rent, insurance, or the loan payment see a dime.

The reason to watch prime cost rather than either half alone: they trade off. A restaurant that cuts labor by understaffing often sees food cost rise from waste and over-portioning. One that tightens food cost with more prep labor moves the cost from one line to the other. The sum is the truth.

Sales by daypart

The fourth number is sales by daypart: breakfast, lunch, dinner, late night, from the POS, weekly. Most restaurants know their dinner is strong. Fewer know that lunch, with a full kitchen and a half-empty room, runs a prime cost of 80% and loses money every day it’s open.

That’s a decision, not a lament. Close for lunch three days a week. Cut lunch to a smaller menu and a smaller crew. Or keep it open on purpose, as marketing for dinner, with the cost written down. Any of those is fine. Not knowing isn’t.

What the four-restaurant cleanup showed

When the 2025 books for four restaurants went from nothing to twelve closed months each, the annual food cost on each one looked ordinary. The monthly view didn’t. On every one, at least two months ran well above the rest, and in each case the owner could name the cause once they saw the month: a vendor price change that stuck, a menu change that didn’t get repriced, a stretch of staffing turnover in the kitchen. Averaged into a year, those months disappeared. Read monthly, they were the year’s biggest controllable cost, and nobody had seen them.

Weekly would have caught them in the week. That’s the whole argument.

What a CFO does with this number

A fractional CFO reads the four numbers together and fixes the right one. Prime cost at 66% with food at 30% and labor at 36% is a scheduling problem. The same 66% with food at 38% and labor at 28% is a purchasing or menu problem. The wrong diagnosis, cutting staff when the issue is the produce contract, makes the next month worse.

Then the CFO turns it into money: a 3-point improvement in food cost on your annual sales is a specific dollar figure, and here are the three menu items and the one vendor it comes from. Or: lunch on Monday through Wednesday costs you this much a year, and here’s what closing it does to the annual number. Owners don’t act on percentages. They act on the dollars.

Where we come in

We close restaurant books monthly with the vendor on every line, food and beverage split at entry, tips handled as tips through in-house payroll, and loans and merchant deposits tied to their statements. The monthly package includes food cost, labor, prime cost, and daypart sales by month. The weekly version runs on top of that for owners who want it, and the count routine is the only part that has to happen in the building.

If your food cost is a number you find out once a year, reach out. We’ll start with a two-hour diagnostic, get the books to monthly, and show you which months you’ve been missing.

Frequently asked questions

What should food cost percentage be?
Typically 28% to 35% of sales depending on concept; the weekly trend matters more than the target.

How often should I calculate it?
Weekly, from purchases and a count, because a yearly average hides the months that spiked.

What is a good prime cost?
Food plus labor under about 60% of sales; over 65% is a structural problem.