A restaurant can be packed every night and still go out of business. The dining room hums, ticket times look fine, the team feels busy—and then the monthly P&L arrives three weeks late and the operator discovers food cost ran six points over target and overtime quietly ate the rest. By then, four payroll runs and twelve produce deliveries have already gone out the door. The money is gone.
This is the single most expensive habit in the restaurant industry: waiting for accounting to tell you what happened last month instead of measuring what is happening this week. The fix is one number, calculated every seven days, and it is called prime cost.
What Prime Cost Actually Measures
Prime cost is the sum of your two most controllable expense lines—cost of goods sold (food and beverage) plus total labor (wages, payroll taxes, benefits, and labor burden)—divided by net sales for the same period.
Everything else on a restaurant P&L—rent, utilities, insurance, marketing, depreciation—is largely fixed in the short run. Prime cost is the part of the business an operator can actually move week to week. That is why seasoned multi-unit operators treat it as the single most important operating metric in the business, ahead of sales, covers, or check average.
The benchmark range that has become industry shorthand:
- Quick-service and fast-casual: roughly 55–60% of net sales
- Full-service and casual dining: roughly 60–62% of net sales
- Fine dining and steakhouse: roughly 60–65% of net sales
Anything north of 65% is a structural problem. Anything south of 55% is unusual and usually means the restaurant is either exceptionally efficient or quietly understaffed in a way that will hurt guest experience.
Why Monthly Tracking Is Too Late
Most independent restaurants close their books on a 28- or 30-day cycle and receive financials anywhere from 10 to 25 days after period end. By the time the operator sees that food cost crept from 30% to 34%, six to seven weeks of damage has already accumulated. On a restaurant doing $80,000 a week in sales, four points of food cost is $12,800 a month walking out the back door—gone before anyone knew to look.
A weekly cadence collapses the feedback loop to seven days. If beef ribeye spikes on Tuesday's invoice, the chef can adjust portioning, raise menu price, or 86 a feature by the following Monday. If labor ran 38% instead of 32% last week, the GM can tighten Saturday's schedule before the next weekend. Weekly prime cost is not an accounting exercise. It is the difference between operating the restaurant and being operated by it.
How to Calculate Prime Cost Weekly: A Step-by-Step Walkthrough
The math is simple. The discipline is in collecting the inputs the same way every week.
Step 1: Define the Operating Week
Pick a fixed seven-day window and never change it. Most restaurants use Monday through Sunday because it captures a full weekend, payroll typically closes Sunday night, and inventory counts can happen first thing Monday morning before deliveries arrive. Use the same week for COGS, labor, and sales. Mixing periods is the single most common mistake operators make.
Step 2: Count Inventory at Week-End
Prime cost requires a periodic inventory because COGS = Beginning Inventory + Purchases − Ending Inventory. Counting only at month-end is the equivalent of weighing yourself once a year and being surprised by the result.
The full count does not need to take all night. Most full-service restaurants can complete a weekly count in 60–90 minutes with two people if shelves are organized in count order, units of measure are standardized, and the count sheet matches the order guide. Stick to the high-impact 20% of items that drive 80% of cost: proteins, cheese, seafood, premium produce, liquor, wine by the bottle. Low-value items like dry goods, spices, and disposables can be counted bi-weekly or monthly without distorting the weekly result meaningfully.
Step 3: Capture Purchases by Invoice Date
Add up every food and beverage invoice with a delivery date inside the week. This is where a real accounts payable workflow earns its keep—if invoices sit in a stack on the GM's desk for two weeks, weekly COGS becomes guesswork. Most operators use an invoice capture tool or simply enter purchases into a spreadsheet each day as deliveries arrive.
Separate food, non-alcoholic beverage, beer, wine, and liquor into their own buckets. Aggregating them hides the leak. A restaurant can have on-target overall food cost while bleeding 8 points on liquor pour cost because the bartender is over-pouring or someone is comping friends.
Step 4: Pull Labor by Pay Period, Not Pay Date
Labor cost is wages earned during the week, not wages paid during the week. If payroll runs every two weeks on Friday, the check that hits this week covers the previous two weeks of work. Use the time clock report for hours worked Monday–Sunday, multiply by hourly rates, and add payroll taxes (typically 7.65% FICA plus federal and state unemployment, workers' comp, and any benefits load—usually another 15–25% on top of gross wages depending on state and benefits).
Salaried managers count too. Pro-rate weekly salary into the labor bucket. Tipped wages and tip credit handling vary by state—make sure your labor number reflects the employer-funded portion only, not customer-funded tips passed through to staff.
Step 5: Net Sales, Not Gross
Net sales means gross sales minus comps, voids, promotions, and sales tax. Including sales tax in the denominator artificially deflates prime cost percentage and makes operators think they are doing better than they are. Pull net sales from the POS daily sales report and reconcile to the deposit before plugging it into the formula.
A Concrete Example
A 120-seat neighborhood bistro, Monday through Sunday:
- Beginning inventory (food + beverage): $24,500
- Purchases: $18,200
- Ending inventory: $22,800
- COGS: $24,500 + $18,200 − $22,800 = $19,900
- Hourly wages: $14,800
- Salary (pro-rated): $3,200
- Payroll taxes and benefits load (~22%): $3,960
- Total labor: $21,960
- Net sales: $68,400
- Prime cost: ($19,900 + $21,960) ÷ $68,400 = 61.2%
For a full-service casual concept, 61.2% is healthy. The operator looks at the components: food cost at 29.1% is on target, labor at 32.1% is slightly elevated. Next week she pulls a labor schedule that trims four shifts of Tuesday and Wednesday afternoon coverage. The following week prime cost is 59.8%. That is what weekly tracking buys you.
The Five Leaks Weekly Tracking Surfaces First
When prime cost moves the wrong direction, the cause is almost always one of these five issues. Knowing the pattern shortens the diagnosis.
1. Portion Drift
The new cook is plating a 9-ounce burger because he eyeballs it. The salad person is over-portioning blue cheese crumbles because the recipe card is missing. Each plate gives away $0.30–$0.80, multiplied by 1,200 covers a week. Use plate-level photo standards, scale-weighed proteins, and randomized line checks. A 3% food cost variance often traces to two or three SKUs being over-portioned by ten percent.
2. Theft and Comp Abuse
Bar inventory variance above 2–3% almost always involves giveaways, free pours, or unrung sales. Comps and voids should require manager approval in the POS and show up on a daily exception report. A bartender comping 14% of liquor sales is not generous—he is running a side business at the owner's expense.
3. Spec Creep on Purchasing
Your meat purveyor swapped Choice for Prime "as an upgrade" and your beef cost is up 18%. Your produce vendor substituted hothouse tomatoes when field tomatoes were short and never reverted. Weekly price-per-unit tracking on the top 20 SKUs catches this in days. Without it, the substitution becomes the new baseline and nobody remembers what you used to pay.
4. Overtime Creep
A scheduler who is afraid of being short-staffed builds in two extra cooks on every Saturday "just in case." Two cooks at $22/hour for six unnecessary hours is $264 a week, $13,700 a year per shift, and that is before overtime kicks in. Track scheduled hours vs. actual hours vs. forecasted sales weekly. A simple sales-per-labor-hour (SPLH) target by daypart prevents the slow creep.
5. Menu Mix Shift
Your high-margin chicken dish was 22% of entrée mix last quarter. After a menu redesign, it dropped to 11% and the low-margin short rib jumped to 26%. Cost percentage moves even though nothing operationally changed. Run weekly mix reports and watch contribution margin per dish, not just food cost percentage.
Building the Weekly Routine
The number itself is worthless if nobody is responsible for it. Restaurants that consistently hit their prime cost target tend to follow a similar Monday-morning routine:
- Sunday night: GM closes the books for the week in POS and pulls the net sales total
- Monday 7:00 AM: Inventory team counts proteins, dairy, seafood, and bar
- Monday 9:00 AM: Bookkeeper enters invoices and inventory counts into the prime cost worksheet
- Monday 11:00 AM: Manager meeting—review prior week's prime cost, identify variances above 1.5 points, assign owners and corrective actions for the current week
- Wednesday: Mid-week check on labor pace vs. forecast; cut shifts if running hot
- Friday: Pre-weekend review of inventory levels and order par adjustments
This rhythm is the difference between an operator who knows what is happening in the building and one who finds out what happened thirty days ago.
Bookkeeping That Makes Weekly Tracking Possible
Weekly prime cost is impossible if invoices are entered monthly, payroll is reconciled quarterly, or inventory only moves on the balance sheet at year-end. The accounting setup needs to support the operating cadence.
A few essentials that make the difference:
- Daily sales journal entry from POS into the books, with comps, voids, and sales tax separated
- Invoices entered to AP within 48 hours and coded to the correct COGS sub-account (food, beer, wine, liquor, non-alc, supplies)
- Inventory adjustments booked weekly so COGS on the books matches the operational number
- Payroll allocations that split FOH, BOH, and management labor by department so labor cost can be analyzed by area, not just totaled
Restaurants that treat the books as a once-a-month tax compliance chore can never run a weekly prime cost. Restaurants that treat the books as a real-time management tool can.
When the Numbers Look Wrong
Some weeks the prime cost number will be obviously off—too low to be real or impossibly high. Before chasing operational ghosts, audit the math first:
- Did inventory get counted in the right units? Cases vs. pounds vs. each is the most common error
- Are purchases coded to the correct week? An invoice dated Sunday but entered Monday belongs to last week's COGS
- Is payroll capturing the full burden, or only gross wages?
- Did a large transfer between locations get double-counted?
- Are you using net sales or gross sales? Sales tax in the denominator can swing the percentage 7 points
A weekly prime cost that bounces wildly between 52% and 68% usually means the inputs are unreliable, not that the restaurant is genuinely volatile. Fix the data discipline before re-engineering the operation.
A Word on Multi-Unit Operators
Once a restaurant group has more than three locations, weekly prime cost stops being a single number and becomes a comparison tool. Posting all locations' weekly prime cost side by side on a shared dashboard creates immediate accountability. The store running 64% has to explain to the store running 57% what is going on—and the operator who has solved a problem can teach the operator who has not.
Multi-unit groups that do this well also separate theoretical prime cost (what cost should be based on actual menu mix and recipe specs) from actual prime cost (what cost was). The gap between the two is the dollar amount of waste, theft, and execution failure. Closing that gap is the entire job of the operations director.
Keep Your Restaurant's Numbers Honest
Weekly prime cost lives or dies on the quality of the underlying books. If invoices, payroll, and sales data are clean, timely, and consistently coded, the metric tells the truth and the operator can act on it. If the books are a mess, the metric is fiction and the restaurant flies blind.
Beancount.io gives restaurant operators plain-text accounting that is transparent, version-controlled, and AI-ready—every transaction is a readable line you can audit, diff, and query without waiting for a closing report. Get started for free and build the financial discipline your operating rhythm depends on.