Purchasing and suppliers in numbers: the traditional method against the Masterestaurant method

GROUP 1 · The gap between theoretical and actual cost

Your biggest leak is not in the price list, it sits in the 2 to 5 food cost points separating what your recipe cards say the food sold should have cost from what inventory says actually left the storeroom. Statista puts sector net margin between 3% and 9%, and full-service between 3% and 5%, so three points of gap eat half of a healthy business. On annual sales of 600,000 dollars, every point means 6,000 dollars walking out the back door with no invoice attached, and five points mean 30,000: more than the average single-unit owner earns in a whole year. One decision comes out of this block, break your food cost down by family —meats, dairy, dry goods, beverages— and hunt the gap where it lives, because the average hides it. A supplier who drops your price 2% and raises your waste 9% is raising your cost, whatever the invoice says.

GROUP 2 · Yield, not price: why the cheap kilo costs more

That is the mistake repeating most often in the quotation folders I review: you negotiate with list prices and you pay with plated cost. Between one number and the other fit butchering waste, post-cooking yield, portioning that drifts from the recipe card, and emergency orders bought at counter price. Follow it all the way through: if your loin goes from 8 to 9 dollars a kilo but yields 12% more portions, cost per portion falls, and the dish you thought was a loser becomes the anchor of your menu. Every new quotation gets tested with measured butchering before you sign, not after three months of invoices. Counting inventory once a month gives you twelve chances a year to catch a problem; counting weekly gives you fifty-two, and that difference is worth more than any discount you negotiate. The reason is plain arithmetic: between counts you have no actual cost, you have an estimate, and no defensible purchasing decision rests on an estimate.

GROUP 3 · Counting frequency: the number that decides whether you measure or guess

With net margins of 3% to 9% per Statista, an operator who spots a two-point deviation six weeks late has already lost a full month of profit before reacting. Here is the order that works: count weekly the ten items carrying 70% of your spend, then extend monthly to the rest. The decision here is not buying software, it is fixing the day and hour of your count and never moving them for the chef's mood. Squeezing the supplier returns less than you think because their margin is already thin, while the levers inside your kitchen cost nothing and pay more. Diego F. Parra has run the financial pillar at Masterestaurant for twenty years and the diagnosis repeats itself: cost reduction programs start in the wrong place. The surrounding data confirms it from another angle; Toast reported labor passed 25% of restaurant expenses in 2024, up from 23% in 2021, so the pressure is not coming from purchasing alone and will not be solved there alone.

GROUP 4 · The supplier is your last lever, not your first

Written recipe cards, measured yield, menu engineering and counting frequency cost neither you nor your supplier a cent. Negotiate, of course, but only once your own house is measured, because a discount laid over a dirty process evaporates at the first inventory. Most restaurants that buy badly are micro-businesses with no negotiating power, and that changes the correct strategy entirely. ABRASEL measured in 2024 that 94% of Brazil's bar and restaurant sector are micro-businesses and 65% are individual micro-entrepreneurs, a profile repeated across Latin America and covered by no corporate purchasing manual. If you buy 4,000 dollars of meat a month, you will not move the needle at a distributor billing millions; what does move it is buying the same product on the same day with two other restaurants in your area. And one figure gets forgotten on the other side of the counter: ABRASEL calculated that every 1,000 reais spent in bars and restaurants injects 3,650 into the economy.

GROUP 5 · What size of operation sits behind these numbers

Your decision is to group up or specialize, never to fight one-on-one alone. One food cost point recovered is worth three to seven times its annual value the day you sell the business, and that is the least discussed reason to put purchasing in order. Sofer Advisors documents restaurants selling at an average multiple of 2.80x to 3.65x EBITDA, with fast-casual between 4x and 7x and fine dining at just 2x to 4x. Run the math to the end: 6,000 dollars a year of closed gap on 600,000 in sales turns into 17,000 to 42,000 dollars of enterprise value depending on format. Publicly traded chains operate at after-tax operating margins of 12% to 13% per WhippleWood CPAs, and they get there through purchasing process, not luck. The decision: treat every cost point as capital, not as this month's expense.

The 3 numbers you should tattoo on yourself

Three numbers and one action for each. First, 2 to 5 points: the typical gap between theoretical and actual cost, and the action is to break your food cost down by family this week and measure which of the four takes more than two points. Second, 3% to 9% net margin per Statista —3% to 5% in full-service—, and the action is to translate every cost point into money from your own annual sales before the next negotiation, so you stop arguing over abstract percentages. Third, 25% of expenses in labor in 2024 against 23% in 2021 per Toast, and the action is to stop hunting all your savings in purchasing when the pressure is spread out. Tomorrow, before requesting a single quotation, write down the measured yield of your five most expensive inputs: that sheet of paper beats your price folder. GROUP 1 · The gap between theoretical and actual.

The figures grouped, and what each group decides

Theoretical cost is what your recipe cards say the food sold should have cost; actual is what inventory says left the storeroom. The National Restaurant Association placed food and beverage cost near 33% of sales across the US industry in 2025, and most operations I audit carry between 2 and 5 points of difference between those two numbers. Mini-conclusion: each point of gap on 600,000 dollars of annual sales is 6,000 dollars walking out the back door without an invoice. One decision follows, split food cost by family (meat, dairy, dry goods, beverages) and hunt the gap where it lives, not in the average. GROUP 2 · Purchase price and its volatility. The USDA food producer price index accumulated mid single digit increases through 2025 and the department itself projected uneven rises by family for 2026, with meat and eggs above average. Mini-conclusion: the mistake is not that prices rise, the mistake is finding out three months later.

The figures grouped, and what each group decides — in practice

An owner reviewing the critical basket every 30 days catches the move while there is still room to react, whether by switching supplier, repositioning the dish or adjusting the portion. GROUP 3 · Yield, trim loss and waste. WRAP documented across its work with British restaurants that every pound invested in cutting food waste returned around seven pounds, and the FAO estimates roughly 13% of food is lost between harvest and the point of sale. Mini-conclusion: waste is the only cost line with a return above 500% and no negotiation required. Concrete decision: weigh organic waste for seven straight days, separated by origin (prep trim, plate returns, expired product), and attack whichever weighs most first. GROUP 4 · Terms, cash and the break-even point. An independent restaurant turns food inventory every 10 to 15 days and pays suppliers between 8 and 30 days depending on how it negotiated. Mini-conclusion: if you turn in 12 days and pay at 30, your supplier is financing 18 days of operation for free, and that spread moves the break-even point more than a 2% discount ever will.

The figures grouped, and what each group decides — key points

The decision: ask for terms before you ask for a rebate, because the rebate gets eaten by half a year of inflation while the terms stay in your account. GROUP 5 · Menu engineering as the answer. When an item rises 12%, the traditional reflex is to raise the dish price; the correct financial reflex is to look at contribution margin in money, not percentage. A dish at 35% food cost leaving 14 dollars of margin is a better business than one at 26% leaving 6 dollars, as long as they turn at similar rates. Mini-conclusion: the percentage governs purchasing, the money governs the menu. Decision: cross food cost and popularity for your 20 best sellers, and move the three that leave the most money per unit into prime menu real estate.