Which dishes to remove from the menu to gain profitability: the high food cost myth against what actually drains the till

Which dishes should you cut from the menu to gain profitability?

Cut the dishes whose dollar contribution margin multiplied by monthly turns lands below 1 % of what your top seller delivers, no matter what their food cost looks like.

The arithmetic is simple and it surprises almost everyone: a dish at 34 % cost that leaves 9.80 USD of margin and sells 180 times a month contributes 1,764 USD, while one at 21 % that leaves 6.40 USD and turns 12 times contributes 76.80 USD. That is 23 to 1 in favor of the "expensive" dish. The sector reference range sits between 28 % and 35 % of price, according to the National Restaurant Association Restaurant Operations Report 2025, and that range describes an entire category; it does not tell you which line to delete from your own menu on Monday. Dollar contribution margin is the only thing that pays rent, payroll and utilities; the food cost percentage is for negotiating with suppliers and hunting portioning drift.

Magnitude rules, percentage tags along

Put it in numbers: 22 % of a 14 USD dish is 3.08 USD of cost and 10.92 of margin, and 33 % of a 46 USD dish is 15.18 of cost and 30.82 of margin. Nearly triple the dollars with the worse of the two percentages. A restaurant billing 80,000 USD monthly in food at a 68 % average margin generates 54,400 USD toward fixed costs; drop to 64 % by pulling expensive anchor dishes and you lose 3,200 USD a month without saving a single dollar of fixed cost. Nobody takes a percentage to the bank. As of September 2026, a well-built menu runs on three ranges and each one carries a different job. The entry range, 9 to 16 USD, carries 24 % to 30 % food cost and exists to fill weekday tables: unit margin sits around 7 to 11 USD and it needs high turns to justify the line.

What each menu price range includes?

The middle range, 17 to 29 USD, is the backbone of the till, with cost between 28 % and 34 % and margins of 12 to 20 USD per plate.

The high range, 30 to 55 USD, absorbs 33 % to 38 % food cost —above the theoretical ceiling— and still leaves 20 to 34 USD per unit. That last range is the one most owners trim out of fear of the percentage, and it is precisely the one holding up the average check. Four levers explain almost all price variation on a menu, and you want them quantified before touching a single dish. Volatile raw material comes first: eggs got so expensive in 2025 that Waffle House added a 0.50 USD per-egg surcharge, according to NPR, and a shock of that kind moves a dish's cost by 8 to 15 points. Second is execution complexity, which drags in inventory SKUs, perishable mise en place and training minutes.

The four factors that move price and margin

Third is buying power: changing your order scale moves 3 to 7 points of cost. And fourth is local market pricing, with 93 % of quick-service restaurants raising prices in 2024 per the Oysterlink compilation. Ignore any one of the four and your menu falls out of tune within a quarter. A chef-owner in Bogotá opened his August 2026 sales report with 64 dishes on the menu, convinced his problem was the ceviche, because it cost 33 % and everyone had warned him that above 30 % you bleed money. That ceviche delivered 11,400 USD a year in contribution margin, ranked second in orders, and was the reason people drove across town to sit down there. The real problem lived in the pasta section: fourteen dishes at 21 % food cost that together never reached 38 units a month and forced six exclusive inventory SKUs into the walk-in. Pulling twelve of those fourteen freed cooler space, cut two daily hours of mise en place and did not move sales by a point.

The case of fourteen pastas at 21 % food cost

The ceviche stayed untouched. A dish with a splendid margin that sells eight times a month is not profitable: it is a museum piece taking up a menu line, cooler space and a cook's attention. Diego F. Parra frames it this way in Masterestaurant menu audits: multiply unit margin by units sold, sort the list high to low, then look at where 80 % of total contribution piles up. On menus of 50 to 70 dishes that 80 % usually comes from 14 to 22 items. Anything below the 20th percentile of total contribution is a candidate to go, with one exception I do defend: the dish that makes no money but brings the group —the decent vegetarian option, the kids' plate— stays, because its value sits in the whole table. Only 10 % of restaurants do quality menu engineering, according to Oracle NetSuite; 60 % never do it at all. Suppose you go from 64 dishes to 32 tomorrow.

What happens if you halve the menu?

The first effect shows up not in food cost but in inventory: active SKUs fall 25 % to 40 %, and with them the product that spoils before ever reaching a plate.

The second lands in the kitchen, where training a new cook drops from three weeks to ten days and Saturday peak ticket times shorten because fewer stations compete for the pass. The third is the one that scares owners and rarely happens: lost sales. When the cut follows contribution rather than percentage, guests migrate toward the dishes that remain, because those were what they already ordered. The real risk is not cutting too deep, it is cutting the wrong dish after reading the wrong column. Before you delete a line, try these four plays in order, because a repaired dish beats a dead one. Renegotiate the dish's raw material with consolidated volume first: folding three similar SKUs into one typically moves cost by 4 to 8 points.

How to negotiate and optimize before deleting?

Second, reset the portion weight to what the guest actually finishes; recovering 25 grams of protein on a 28 USD plate hands back roughly 1.60 USD of margin.

Third, relocate the dish on the page, since eyes land on the top third of every section and turns rise there. Fourth, raise the price by 1.50 USD and measure four weeks, because 42 % of restaurants already pushed increases through in 2023 according to Toast and guests absorbed them. If contribution still sits at the bottom of the list after that cycle, then yes, delete it on Monday. Magnitude first, percentage second. Contribution margin is selling price minus raw material cost per portion, and it is the only figure that pays rent, payroll and utilities. A venue billing 80,000 USD monthly in food with a 68 % average contribution margin generates 54,400 USD toward fixed costs; drop to 64 % by removing expensive anchor items and it loses 3,200 USD a month without saving a single dollar of fixed cost.

Four differences that decide the menu

Food cost percentage belongs in supplier negotiations and portioning audits, not in menu-cutting decisions. Turnover weights everything else. An item with a beautiful margin selling eight times a month is not profitable, it is a museum piece taking up a menu line, walk-in space and a cook's attention. Our operating threshold is 70 % of average category turnover: below that, an item gets 60 days to climb or it goes. And when it goes, sales do not evaporate, they migrate — 55 to 75 % of that demand moves to the adjacent reference, almost always a better-margin one. Complexity carries a price even when no recipe card shows it. Count how many inventory references exist ONLY because of one item, how many mise en place minutes it eats each service, and how much waste it generates on the days it does not sell.

Four differences that decide the menu — in practice

In a 12-person kitchen with 75 % annual turnover, each extra menu item costs roughly 40 training minutes per new hire per year; with nine hires a year and a loaded labour cost of 9 USD per hour, that is 54 USD annually per item just to teach it. Multiply by twenty items nobody orders. Removal is the last resort, never the first. Out of every ten items the analysis flags red, three actually leave, four get repriced, two get reformulated around a neighbouring item's inputs to kill the exclusive reference, and one simply moves position on the printed menu and comes back to life. According to Gregg Rapp, the menu engineering consultant regularly quoted by US industry press, design and placement within the menu measurably shift what guests order, which makes redesign a cheaper lever than deletion. Cutting first and thinking later is what sinks average checks.