Masterestaurant Recipe-Cost Gap Index 2026: the leak between theoretical and served cost

Finding 1 — How far is your real food cost from the theoretical one?

The average gap between theoretical recipe cost and cost served is 4.7 food-cost points. Across 8,400 accounts audited over three years, the average recipe card says 28.3%, but the real cost walking out the door is 33.0%.

Those 4.7 points aren't an accounting error: they're pure margin leakage. And they don't show up in your spreadsheet because your spreadsheet trusts the recipe. In a venue billing €900,000 a year with a 28% target food cost, that gap means roughly €42,000 vanishing with no invoice attached. The number that matters isn't the sector average: it's your own class of operation. A 40-cover gastrobar and a six-venue group don't play the same league, and comparing yourself against a flat figure blinds you exactly where the money leaks. Cleaning and trimming waste devours between 1.2 and 2.1 food-cost points that the recipe ignores completely.

Finding 2 — Real product yield eats up to 2.1 points

The recipe card for that tenderloin assumes you buy 1 kilo and plate 1 kilo. Reality: from a 3-kilo loin, after degreasing, removing the sinew and trimming, you're left with 2.1 usable kilos. That 30% waste isn't on the card, but it is on your supplier invoice. Same with fish: a whole sea bass yields 45-52% in clean fillet, not the 100% the market price charges. Diego F. Parra has seen it in dozens of kitchens: the recipe cost is built on purchased weight, never on real yield after butchering. Fixing just this —measuring yield piece by piece— recovers an average of 1.6 points in Masterestaurant audits. Without controlled gram weights, the real portion exceeds the recipe by 8-14% on average, and this is the largest leak of all. The card says 180 grams of protein; the cook on a Saturday at 22:40 plates 205 because he goes by eye and prefers the dish to look full.

Finding 3 — Over-portioning is the biggest leak and the quietest

Multiply those extra 25 grams by 180 plates in a service and by 300 services a year: it's tons of product given away that nobody counts. It's the quietest leak because there's no theft, no expired stock, no wrong invoice: every plate simply weighs more than you charge for. In audits, putting a scale on the line and standardizing gram weights cut food cost by 1.5 to 2.8 points in the first six weeks, without touching prices or suppliers. It's the most profitable adjustment and the one kitchens resist most. Off-recipe purchases —supplier substitutions, different formats and stale prices— add between 0.9 and 1.6 points to the gap. Your recipe cost was calculated with oil at €4.20 a liter and the usual supplier. But last month that oil jumped to €6.80 and nobody touched the card; the sous chef bought a different cheese brand because it ran out, at 12% more cost; and the 5-kilo format that throws off the unit price came in unrecorded.

Finding 4 — Buying off-recipe adds up to 1.6 points you don't see

Each substitution seems minor, but together they move the needle. In a six-venue group we audited, 34% of the SKUs had a price out of date by more than 90 days. Reconciling purchases against sales every month, SKU by SKU, is tedious, but it's exactly what separates knowing your food cost from believing it. Expired stock, breakage, unrecorded staff meals and pilferage add between 0.8 and 1.4 points depending on the stock control you keep. This block is the thermometer of your warehouse discipline: where inventory is counted by hand every week, the figure stays at 0.8; where nobody counts and everything is estimated, it spikes to 1.4 and beyond. The expired item tossed unrecorded, the case of glasses that breaks, the staff eating without logging consumption, the shrink that camouflages theft: none of these lines has an invoice to claim against, and that's why they live in the shadow of the recipe cost.

Finding 5 — Operational waste and pilferage: 1.4 points gone without a sound

Control isn't suspecting your team; it's measuring. A weekly inventory counted and cross-checked against theoretical outflows turns those diffuse 1.4 points into a concrete figure you can attack. This benchmark puts a figure on the gap by segment and size so you stop comparing yourself against a useless average. The 4.7 points are the mean of the 8,400 accounts, but the real range runs from 2.9 points in operations with a scale, weekly inventory and living recipe cards, up to 7.8 points in venues that work from memory. The rule is simple: if your gap exceeds 4.7 points, you don't have a purchase-price problem, you have a control problem. The order of attack is also clear by leak size: gram weights first (up to 2.8 points recoverable), then yield by butchering (1.6), next purchase reconciliation (1.3) and finally inventory to close the waste.

Finding 6 — How to read this index for your own class of operation

Diego F. Parra insists at Masterestaurant: don't chase the perfect theoretical food cost, close the gap with the real one. That's where the money is. Real product yield: trim and cleaning waste eats between 1.2 and 2.1 points the spec sheet ignores entirely. Over-portioning: without controlled grammage, the real portion exceeds the spec by 8-14% on average; the quietest leak and the biggest. Off-recipe purchasing: supplier swaps, different formats and outdated prices add 0.9-1.6 points. Operational waste and petty theft: expired stock, breakage, uncounted staff meals and shrinkage add 0.8 to 1.4 points depending on inventory control.