How to make a restaurant profitable: the data behind the traditional method versus the Masterestaurant method

Prime cost below 60% is the only number that decides whether you make money

A restaurant becomes profitable when raw materials plus total payroll fall below 60% of net sales, and that figure gets reviewed every Monday, not on the 8th of the following month. The National Restaurant Association places the median operating margin for full service between 3% and 5% of sales, a cushion so thin that three points of drift in purchasing erase the whole year's profit. In that 92-seat place in a corporate district, food cost went from 30.4% to 36.1% over eleven weeks without anyone noticing; against 74,000 dollars in monthly sales, those 5.7 points amount to roughly 4,200 dollars a month walking out the back door. Sales, meanwhile, were growing 14% year over year. That is the trap: growing and getting poorer at the same time. Measuring on Monday against the closed week lets you correct course; measuring on the 10th of the following month only lets you grieve.

Why measuring on Monday changes the outcome and measuring on the 10th does not?

The difference between those two calendars lies not in the formula, which is identical, but in how many decisions remain alive when the number shows up.

With a weekly figure you can still switch protein suppliers, fix a portion that crept from 180 to 210 grams, or pull a dish whose cost exploded because 80-90% ground beef moved from 4.56 to 5.63 dollars per pound according to USDA price data. With a monthly figure that variance has already multiplied across four weeks of service, the money has already reached the supplier and the inventory has already been consumed. The same calculation, applied 30 days late, stops being management control and becomes accounting archaeology. Spreading payroll into the cost of each dish looks like accounting prudence and is the mistake that destroys the most profit in this business. When you allocate 23 or 27 points of overhead across every unit sold, a dish with a real food cost of 29% shows up on the sheet at 55%, and the owner reacts by raising prices until the demand he already had walks away.

Payroll does not belong in the plate: it belongs in break-even

The cook's wage does not change when plate number 300 leaves the pass; the chicken does. That is why Masterestaurant always separates the two natures: raw material to the plate, structure to break-even. Diego F. Parra has repeated it for twenty years with the same dry line: show me the percentage, not the total. The operating ceiling for raw material per dish is 32%, and that 32% is a MAXIMUM that already signals a menu engineering problem, not a target to reach for. Reference percentages do not apply the same way in a 40-seat room as in a six-unit group, and that deserves saying before someone copies a benchmark that belongs to another business. In a small restaurant, say 30,000 dollars in monthly sales, one point of food cost is 300 dollars: there the real lever is portioning and waste, because the 4% to 10% of inventory an average operation throws away according to The Restaurant HQ weighs more than any price negotiation.

How to read these numbers in YOUR operation: three scenarios?

In a mid-size unit doing 70,000 to 120,000, one point is worth between 700 and 1,200 dollars a month and a full weekly inventory with blind counts already pays for itself.

In a group of three or more locations, the leak hides in the variance BETWEEN units: two kitchens with the same menu and three points of difference mean one of them has a process problem, not a market problem. The figures I use here come from verifiable public sources, and their limits deserve stating before anyone treats them as law. The operating margin ranges and the 8.8% menu inflation of March 2023, the highest in more than two decades, come from the National Restaurant Association and describe the United States market; input prices come from the USDA Economic Research Service, with farm-level eggs up 43.1% in 2024; the waste data comes from ReFED, which attributes 17.9% of the country's 2024 food surplus to foodservice and more than 43% of that surplus to full service restaurants.

Where these benchmarks come from and what they do NOT tell you?

None of those sources knows your menu, your rent or your channel mix. They help you size the problem and argue with a supplier;

they do not replace the inventory in your kitchen, which is the only figure that describes your business. Between 4% and 10% of the inventory a restaurant buys ends up in the trash according to The Restaurant HQ, and that range, applied to 25,000 dollars in monthly purchasing, means somewhere between 1,000 and 2,500 dollars you paid for, received, stored and threw out. ReFED calculates that foodservice generated 17.9% of the American food surplus in 2024, with full service restaurants contributing more than 43% of that total: waste is not an accident of the sector, it is its structure. I got this wrong for years, because I attacked waste with posters and awareness talks when the fix was boringly physical: written spec sheets with exact grammage, a scale on the line, and residue counted by category for two weeks.

Waste is the leak nobody invoices and everybody pays

The first time a kitchen weighs its own garbage the number stings; by the second week it has already dropped. Suppose you apply the industry's favorite recipe and lift the average check by 8% through upselling and a menu redesign, leaving the cost structure untouched. On 74,000 dollars a month your sales rise to 79,920, some 5,900 additional dollars. But if prime cost stays at 68%, then 4,012 of those 5,900 disappear into raw material and payroll, and 1,888 dollars reach the contribution line, against which rent, utilities and financing still run. Now flip the exercise: bring prime cost from 68% down to 60% without selling a single dollar more and you release 5,920 dollars a month out of the sales you ALREADY have, without training anyone in selling techniques or risking a guest who felt pressured. A place doing 74,000 in sales at 68% prime cost leaves less cash than one doing 58,000 at 57%, even though the first one sounds better in any hallway conversation.

Start Monday: the four-figure routine that holds the margin

The whole system fits on one sheet and takes 40 minutes every Monday morning. Close a physical inventory of the ten families that represent 80% of your purchasing, record net sales for the week, add the payroll for those seven days with taxes included, and calculate the two percentages: food cost and prime cost. If prime cost exceeds 62%, nothing gets bought off-list that week and you review the three families with the largest variance against the prior week. Mexico's restaurant industry accounts for 12.2% of the country's economic units according to INEGI and CANIRAC, and the vast majority of those businesses operate without this figure in hand, which explains the closure rates rather well. Forty minutes a week against four thousand dollars a month in leakage: the arithmetic of that decision does not invite debate. The bigger difference is not what gets measured but when.

Where profitability is actually decided?

A food cost figure produced on the 10th of the following month describes money already spent, already collected and already banked;

a food cost figure produced on Monday over the closed week still allows you to swap a supplier, adjust a portion or pull a dish before the variance multiplies by four weeks. Loading labor into plate cost looks prudent and is the accounting mistake that destroys the most profit in this industry. Spread 23 or 27 points of structure across every unit sold and the plate appears to cost 55%, so the owner raises prices until demand walks away, when labor does not move with the 300th plate of the day. Labor belongs in the break-even, always. The traditional method asks how much we sold. The Masterestaurant method asks how much stayed per dollar that came in, and which line took the rest. Two different questions, two different businesses: one grows revenue with flat cash, the other grows cash even when revenue barely moves.

Where profitability is actually decided — in practice?

Menu engineering: most menus carry between four and seven dishes that sell heavily and leave little, and that intersection only becomes visible when you plot contribution margin in dollars on one axis and units sold on the other.

Trimming a menu without that matrix is guesswork; with the matrix, it is a supported decision. If your restaurant runs a digital menu, Masterestaurant ALWAYS recommends keeping the physical menu as well. The printed menu governs service pacing, menu narrative and suggestive selling, which is where a check with margin gets built; the QR complements it with delivery, accessibility, price updates and analytics. Both, each in its own role, never one alone.