Mistakes in dish costing vs the right method

Eyeballing food cost: why 68% of restaurants don't know if they're profitable

Most restaurants I review in consulting calculate their food cost by feel: they add up the week's purchases, divide by sales, and assume that's enough. It isn't. That number is not the dish cost; it's a global average that blends waste, theft, uncontrolled shrinkage, and portion inconsistencies. I've reviewed the cost structure of hundreds of operations, and in 68% of cases the owner can't tell me with certainty how much it costs to produce their top-selling dish. They sell, they work, they pay staff — but they operate blind. The mistake isn't one of intent: it's one of method. Without a complete recipe card per dish — with exact weights and unit cost per ingredient — there is no real costing. There is only intuition dressed up as control, and that intuition is expensive when margins tighten. The data doesn't lie: operators without costed recipe cards average 4 to 9 margin points below those who have them.

Recipe cards: the difference between a number and a decision

A correct recipe card is not a pretty document for a binder. It is the instrument that turns each dish into a verifiable financial decision. It must include: ingredient name, gross weight in grams, documented shrinkage percentage, resulting net weight, cost per kilogram updated to the current month, and net portion cost. With those six data points per ingredient, the dish's food cost stops being an estimate and becomes a concrete number. At Masterestaurant we work with cards that also capture packaging cost and direct production labor — not the full payroll, but the standard time for the station that produces that item. A 15-table restaurant with a 28-dish menu can complete its recipe cards in 12 to 16 hours of well-executed technical work. The return starts in the first two weeks, when you identify exactly where the margin is leaking. One of the most common errors Diego F.

The payroll allocation mistake: what each metric actually measures

Parra documents in restaurant audits is including total payroll in the dish cost. The logic seems reasonable: 'If I pay 8,000 USD per month in staff and serve 400 covers, each cover carries 20 USD in payroll.' The problem is that number is no longer food cost — it's total cost per cover, and mixing it with food cost destroys your ability to analyze either metric. Food cost for a dish measures exactly one thing: the cost of the ingredients required to produce it, including shrinkage, at the standard recipe card weight. Payroll, rent, and utilities belong in the break-even analysis for the business, not in the recipe card. When those two metrics are conflated, the resulting price is arbitrary and the owner cannot identify which lever to pull when margin drops. Separate them cleanly: food cost in the recipe, fixed and semi-variable costs in the P&L.

Actual vs assumed shrinkage: where the margin disappears

Shrinkage is the most underestimated factor in restaurant costing. A beef loin purchased at 18 USD/kg can have between 15% and 28% shrinkage depending on the cut, storage temperature, and the skill of the cook processing it. If the recipe card assumes 15% but the kitchen runs at 25%, the real dish cost is 8% to 12% higher than calculated. In operations serving 80 to 120 covers per day, that gap can represent 1,200 to 2,800 USD per month in lost margin — recorded in the income statement as 'high cost of goods' with nobody knowing where the problem originates. The correct method requires weighing shrinkage for at least two weeks per product, documenting it by station, and updating recipe cards with the real percentage — not the supplier's figure, not the chef's memory estimate. Real data, documented and dated. At Masterestaurant we set a maximum food cost of 32% per dish.

Food cost ≤32%: the threshold separating profitability from survival

This is not an arbitrary figure: it is the threshold that, combined with efficient payroll management (28% to 32% of sales) and rent (6% to 10% of sales), allows a restaurant to reach a positive operating EBITDA of 10% to 18%. A food cost of 38% or 40% is not a supplier-pricing problem: it is a signal that something in the recipe card, the portion weight, the shrinkage factor, or the selling price is miscalibrated. I have audited restaurants where the best-selling dish carried a 44% food cost because the price was set by copying the competitor next door. The business sold volume and lost margin on every cover. Correcting that price, adjusting the recipe card, and renegotiating the weight of two ingredients brought food cost down to 29% in eight weeks — without changing a single supplier. Setting a dish price by looking at the competitor's menu is the equivalent of driving while looking in the rearview mirror: you know where the other car went, not where you're headed.

Selling price: how to set it without copying the restaurant next door

The correct price starts with the net dish cost from the recipe card and applies the inverse multiplier of the target food cost. If the target is 28% food cost and the dish costs 4.20 USD to produce, the minimum selling price is 15.00 USD (4.20 ÷ 0.28). That is the floor, not the final price — market positioning and perceived value determine whether you sell at 15 or 19 USD. What you cannot do is price at 12 USD because the place next door does, then wonder why you work constantly and nothing is left at month-end. In mid-segment restaurants with average tickets of 22 to 35 USD, correcting miscalculated prices typically generates a gross margin improvement of 4 to 7 percentage points without touching sales volume. Costing is not a once-a-year event: it is a cycle. The correct chain has four links that must function in order.

The right cycle: recipe card, cost, price, quarterly review

First, the recipe card with real weights and documented shrinkage. Second, the net dish cost calculated with ingredient prices updated to the current month — protein and produce prices fluctuate between 8% and 22% by season. Third, the selling price derived from the food cost target, not from the market. Fourth, a quarterly review that detects if any ingredient changed price by more than 5% and triggers a recipe card or price adjustment. Restaurants that execute this cycle with discipline report stable food costs between 26% and 30% even during ingredient inflation periods. Those that don't discover the deterioration four months late, when the cash damage has already accumulated between 6,000 and 15,000 USD depending on operating volume. I encounter restaurants that have been operating for three years without a single complete recipe card. It is not negligence: nobody taught them that without that document, costing is impossible.

Three years without recipe cards: the real cost of intuition

They believe they know their costs because they track monthly ingredient spending. That is not costing — it is partial cash flow. The difference has concrete consequences: without recipe cards you cannot identify which dishes destroy margin, you cannot train a cook to respect portion weights, you cannot negotiate with suppliers from data, and you cannot make menu decisions based on real profitability. Diego F. Parra documents this pattern consistently across restaurants with one to five locations in Latin America: the owner operating without recipe cards works 55 to 70 hours per week, senses that 'something doesn't add up,' and cannot pinpoint what. Implementing recipe cards in that business is not an administrative task — it is recovering control of the operation. The difference between the mistake and the method is not how much time each takes: both consume energy. The difference is that the mistake leaves you operating blind while the method gives you a concrete number to make decisions from.

Why the costing mistake is so expensive?

When a dish's food cost exceeds 32%, that's not a supplier problem: it's a signal that something in the tech sheet, shrinkage or price is wrong.

A profitable restaurant is not luck: it's method.