What Is Prime Cost? Formula, Benchmarks, and How to Lower It

Restaurant prime cost is the total of your cost of goods sold (the food and beverage you used) and your total labor cost (wages plus payroll taxes and benefits) for the same period, usually shown as a percentage of sales. It is the largest block of spending a restaurant can control week to week, which is why operators watch it more closely than any other number. The two decisions that matter are what you count on each side, and how you calculate it weekly when you don't do a full inventory count every week.

Prime cost is the first number to get right in the guide to reducing restaurant costs without hurting service. The National Restaurant Association's 2025 operations data shows why: the median limited-service restaurant spent 65 cents of every sales dollar on food, beverage and labor in 2024, while the median limited-service operator kept 4.0 percent of sales as pre-tax income. When two cost lines take nearly two-thirds of every dollar, small moves in either one decide whether the year is profitable.

The prime cost formula

Formula: Prime cost = cost of goods sold (COGS) + total labor cost. Prime cost percentage = prime cost ÷ total sales × 100. Use the same period, a week or a month, for all three numbers.

COGS is what you used, not what you bought, so it needs an inventory count: COGS = beginning inventory + purchases − ending inventory. Total labor is everything you spent on people in the period, not only the hourly payroll.

Worked example: Say a restaurant does $40,000 in sales in a week. Food cost is $11,200 and beverage cost is $1,600, so COGS is $12,800. Wages come to $11,000, and payroll taxes, workers' comp, health insurance and other benefits add $2,200, so total labor is $13,200. Prime cost is $12,800 + $13,200 = $26,000, or 65.0% of sales. That leaves $14,000, or 35% of sales, to cover rent, utilities, card fees, repairs, marketing and profit.

Now run the same week the way many operators do, with wages only: $12,800 + $11,000 = $23,800, or 59.5% of sales. The restaurant looks comfortably inside the usual targets when it's really at the top of them. In this example the gap is 5.5 points, or $2,200 a week, which comes to $114,400 over 52 weeks. Count all of labor, every time.

Run your own numbers below. The prime cost calculator shows prime cost in dollars and as a share of sales, with food and labor broken out, and the dollar gap to your own target if you enter one.

What counts as cost of goods sold and labor

Most disagreements about prime cost come from what goes into it. The table shows the common treatment. Where practice varies, pick one treatment, write it down and keep it, so this month compares with last month and one location compares with another.

Why payroll taxes and benefits matter so much

Restaurant365's definition of labor cost is "the total wages your employees have earned during that specific period of time, payroll taxes, benefits, and insurance." The taxes alone are not small. Per the IRS, the employer pays 6.2% of wages for Social Security and 1.45% for Medicare, before federal and state unemployment taxes.

Benefits add more. In the Bureau of Labor Statistics' June 2026 compensation data, private employers in leisure and hospitality paid $17.25 an hour in wages and salaries and $4.08 in benefits, so benefits were 19.1% of total compensation cost. That's roughly 24 cents of benefits for every wage dollar. The worked example above used $2,200 on $11,000 of wages, or 20 cents, so if anything it understates the gap.

What is a good prime cost percentage?

The ranges you'll see quoted are targets, and they come from software companies, not from survey data. The National Restaurant Association and accounting firms publish what operators actually report. Both are useful, as long as you know which is which.

Three things stand out. First, the median limited-service operator reported a prime cost at the very top of the target ranges, not in the middle of them. Second, the NRA's public summaries don't give a full-service prime cost median, but its separate full-service medians point the same way: a restaurant at both would run about 68.5% before alcohol cost, above the 60 to 65% target. Medians of separate cost lines don't add up to a median prime cost, so treat that as a rough read, not a statistic.

Third, prime cost tracks profit. Citrin Cooperman sorted its full-service clients by EBITDA margin. In its 2025 benchmarking report, top performers (EBITDA of 17.5% of revenue or more) spent 23.12% of revenue on food, beverage and disposables and 31.40% on labor including taxes and benefits, a prime cost of 54.52%. Mid-range performers ran 25.54% and 36.44%, or 61.98%. Low performers (EBITDA under 8.5%) ran 27.69% and 40.88%, or 68.57%. Note that Citrin counts disposables in its cost of goods, and its sample is larger full-service restaurants.

The NRA itself says its figures are "not intended to represent standards or goals for individual restaurants." A steakhouse and a pizza counter can both be healthy at very different prime costs. Use the ranges as a sanity check, and judge yourself mainly against your own trend.

How often should you calculate prime cost?

Weekly. A monthly number arrives after four weeks of orders and schedules are already spent. 7shifts calls a weekly prime cost "your core rhythm," and Restaurant365 warns that calculating it only once a quarter or once a year misses weekly chances to trim food and labor cost. A routine that works for most single-unit and small multi-unit operators:

1. Daily: labor against sales. Pull hours and labor dollars from your timeclock or POS each morning and compare them with yesterday's sales. Labor is the half you can still change this week.
2. Weekly: estimated prime cost. Count your highest-cost items (proteins, cheese, alcohol, anything you buy by the case at a high price) and estimate the rest, as described below. Add total labor, including an estimate of payroll taxes and benefits. Label the result an estimate.
3. Monthly: reconcile with a full count. Count everything, calculate actual COGS for the month, and compare it with the sum of your weekly estimates. If they differ by more than a point of sales, find out why before you trust the weekly trend.
4. At every menu or vendor change: recost. A price increase from a supplier moves your food cost the day it lands, not at month end.

Set a target from your own best recent weeks, not a number from the internet, and move it a point at a time.

Weekly prime cost from a partial count

Counting only your high-cost items each week is a sensible shortcut, but it changes what the number means. If you pair a count of a few items with all of the week's purchases, you are measuring usage for some items and buying for the rest, and anything that built up or ran down on the uncounted shelves lands in your COGS. The result is not actual food cost. Treat each part differently:

• Counted items: usage = beginning count + purchases of those items − ending count.
• Uncounted categories (produce, dry goods, dairy, paper if you include it): use that week's purchases as usage, which assumes their stock level didn't change. When you know it did, such as a monthly stock-up of dry goods, use the share of sales those categories ran at in your last full count instead.
• Estimated COGS is the two added together. Keep the categories and dates the same every week, so this week compares with last.

Worked example (an estimate, not actual COGS): Say a week's sales are $40,000. The counted items start at $4,000, take $6,500 in deliveries and end at $3,800, so $6,700 was used. Uncounted categories had $7,400 in purchases, including a $1,500 monthly stock-up of dry goods and paper. Adding all of it would put COGS at $14,100, or 35.3% of sales. Instead, say last month's full count showed the uncounted categories running at 14% of sales: $5,600 this week. Estimated COGS is $6,700 + $5,600 = $12,300, or 30.8% of sales. The month-end full count tells you whether that 14% still holds.

When the month's full count and the weekly estimates disagree, the usual causes are an uncounted category whose stock moved, an invoice dated in the wrong week, or an item counted at the start of the month but not the end. Fix the method, not just the number.

How to lower prime cost without hurting service

Prime cost has two halves, and each has levers the guest never sees. Work on those first.

Food and beverage cost:

• Recost your top sellers against current invoices, and reprice or re-spec the ones whose cost has climbed.
• Compare vendor prices on your biggest-spend items every month, not once a year.
• Portion with scales and spoons sized to the recipe, so the plate matches the cost card.
• Compare theoretical food cost (what your recipes say you should have used) with actual food cost from a full count. Before you put the gap down to waste, over-portioning or theft, check for counting errors, missing invoices and unlogged comps.
• Drop or replace dishes that sell poorly and cost too much to make.

The guide to reducing food cost in a restaurant covers these steps in detail, and the food cost calculator works out plate cost and food cost percentage.

Labor cost:

• Schedule to forecast demand by daypart (lunch, dinner, late night), not to habit.
• Catch overtime before the week ends, not on payroll day.
• Cross-train so a slow shift can run with fewer people without gaps.
• Keep good people. Every replacement costs money to hire and train.

The guide to reducing labor cost in a restaurant walks through each lever, including what not to cut.

What not to cut: staffing at peak hours, portions on signature dishes, training and food safety. Those cuts show up in ticket times, reviews and repeat visits within weeks, and lost sales raise prime cost as a percentage even when the dollars fall.

The tracking is the hard part for a busy operator. Kutlerri.ai's Prime Cost Optimization Agent, currently in beta, tracks food and labor cost, flags vendor price increases, prices menu items off what they cost to make, recommends recipes to drop or replace, and matches labor to demand.

Prime cost vs food cost vs COGS

These three get mixed up, and the difference matters.

• Food cost percentage is the cost of food used in a period ÷ food sales for the same period. It covers ingredients only.
• COGS adds beverages, and in some charts of accounts paper and packaging.
• Prime cost adds total labor to COGS.

A restaurant can hit a textbook food cost and still lose money if labor has drifted, which is why prime cost, not food cost, is the number to manage weekly. For how prime cost fits with occupancy, delivery commission, utilities and the rest of the cost stack, see the guide to reducing restaurant costs.