Contribution Margin vs Food Cost Percentage (Why Food Cost % Lies)

The mistake I see most often in menu pricing? Operators chasing food cost percentage like it’s the score that matters. They see a 38% food cost dish and figure it’s bleeding them. They see a 22% food cost dish and figure it’s carrying the place. Both reads are wrong, and they’re wrong in a way that costs real money. Food cost % is a ratio. Profit gets paid in dollars, and the dollars don’t care what the ratio says.

Here’s the contribution margin argument with the math actually worked out, so you don’t have to take my word for it.

The two numbers #

Food cost percentage is the share of revenue from each item that goes to ingredient cost.

Food Cost % = Food Cost ÷ Menu Price × 100

Contribution margin (in dollars) is the dollar amount that each item contributes to covering fixed costs and profit.

Contribution Margin = Menu Price − Variable Food Cost

Same data, different framing. Both have a role. They just answer different questions, and grabbing the wrong one when you’re making a menu call is how you talk yourself into the dumb decision.

The classic example #

A $14 salad with 22% food cost vs a $42 steak with 38% food cost.

Item	Price	Food Cost $	Food Cost %	CM $	CM %
Salad	$14	$3.08	22%	$10.92	78%
Steak	$42	$15.96	38%	$26.04	62%

By food cost percentage, the salad wins. 22% beats 38%, end of story. So the operator features the salad, buries the steak, and starts wondering out loud whether the steak even earns its spot on the menu anymore.

By contribution margin in dollars, the steak wins by a mile. $26.04 per cover vs $10.92. That’s 2.4x more money on every plate that walks out.

So which one’s right? Dollars. It’s not close. Your landlord, your line cooks, and your produce guy all get paid in dollars, not percentages, and none of them will take 78% of a salad. The ratio is just a sanity check that you’re earning those dollars without waste. A 38% food cost throwing off $26 of contribution per cover is pulling way more weight than a 22% food cost throwing off $11.

When the percentage trap costs you money #

The trap shows up in three operational decisions.

1. Menu placement. Operators feature their low-food-cost items at the top of the menu. The math says feature the high-CM-dollar items, even if they have higher food cost percentages.

2. Sales mix shift decisions. Food cost runs hot, so the move is to push people toward the low-food-cost stuff. But if that push moves a customer off the $42 steak (CM $26.04) and onto the $14 salad (CM $10.92), you just handed back $15.12 in margin to make a report look prettier. Five switchers a night is $75 of margin gone every night, which is $27,000 a year. The food cost line on the report drops. The bottom line drops with it.

3. Item removal decisions. A high-food-cost item that actually sells (a Plowhorse, in menu engineering terms) gets axed because “the food cost is too high.” Say the killed item threw off $14 of CM per cover and whatever replaces it throws off $9. Now every customer you kept is $5 a cover less profitable than they were last week, and you did that to yourself.

So the pattern is simple. Food cost percentage tells you how efficient you are. CM dollars tell you how much money you actually keep. Chase the money.

The exception: when ratio matters #

There’s one place food cost percentage beats CM dollars, and it’s tracking your whole P&L against industry benchmarks.

Total food cost % is the number that lets you line your operation up against everyone else’s. A casual full-service spot running 35% food cost is running hot. The same spot at 29% is healthy. But that comparison only holds at the operational level. The second you drag it down to the per-dish level, it lies to you.

So:

• Operational level (whole P&L): Track food cost % against benchmarks. This is the operational health check.
• Menu decisions (per item): Optimize for CM dollars. This is where profit gets made or lost.

Operators mash the two together and start using the whole-P&L signal to make individual menu calls. That’s the trap, every single time.

A worked example: same revenue, different mix #

Two scenarios, same restaurant, same monthly revenue ($120,000).

Scenario A: Heavy on low-food-cost items

• 60% of revenue from salads, sandwiches, apps (avg CM 80%, avg ticket $13)
• 40% of revenue from entrées and steaks (avg CM 65%, avg ticket $36)
• Total food cost: 28%
• Total contribution margin: $120,000 × 72% (1-28%) = $86,400

Scenario B: Heavy on high-food-cost-but-high-CM-dollar items

• 40% of revenue from salads, sandwiches, apps
• 60% of revenue from entrées and steaks
• Total food cost: 33% (higher because entrées have higher food cost %)
• Total contribution margin: $120,000 × 67% = $80,400

Hold on. Scenario A wins this one. That 5-point food cost difference costs Scenario B $6,000 in CM. So food cost percentage does matter at the aggregate level after all, and if I stopped here I’d be telling you the opposite of the whole point.

But I left something out: average ticket. Scenario B’s mix doesn’t just shift the percentages, it puts a different check on the table. Push that 60-40 entree-heavy mix and the average check climbs from $32 to $38. Now the revenue isn’t $120K, it’s more like $138K, and at 67% CM that’s $92,460 of contribution margin. Scenario B wins by $6,000.

That’s the thing percentage framing misses cold, the average-check effect. The real analysis is contribution margin in dollars across your whole sales mix. Not “lower food cost % = better,” which is the bumper sticker that gets people in trouble.

The right way to use both #

Here’s the discipline.

Per-item menu decisions: Use CM in dollars. Feature the high-CM-dollar items, put them at the top of the menu where eyes land, and train your servers to upsell them. And don’t you dare kill an item because its food cost % is “too high.” Kill it only when its CM dollars fall below the category average. That’s the only number that earns an item a death sentence.

Aggregate operational tracking: Use food cost % against benchmark. If your weekly food cost runs 5 points above your segment median, that’s your signal something’s off. Go hunt the variance, recipe compliance, portion drift, pricing. The answer is almost never “I should sell more salads.”

Menu engineering: Use CM dollars on one axis and velocity on the other (see What Is Menu Engineering). This is the dollar-framing applied systematically across the menu.

Pricing decisions: Use both. A new menu item should have CM dollars that beat the category average AND food cost % that doesn’t pull the aggregate above benchmark.

What this looks like in the calculator #

The menu pricing calculator on this site lets you set target prices by the food cost % method or the CM-dollar method. Toggle between them and watch the numbers move. Most operators are genuinely surprised how far apart the recommended prices land depending on which framing they trust.

The menu engineering calculator plots CM in dollars on the Y axis, not CM %. That’s the version that hands you the right action list instead of a pretty one.

What to do today #

Pull last month’s top 10 items by sales count. Work out CM in dollars for each one, then sort by CM dollars, high to low. Now sort that exact same list by food cost %, low to high. If you’re doing it right, the two lists won’t match. The items riding high on the food-cost-% list but sitting low on the CM-dollars list? Those are the ones you’ve been over-featuring this whole time. The items high on CM dollars but lower on food cost %? Those are the ones you’ve been hiding.

And the fix isn’t a recipe overhaul or a vendor renegotiation. It’s where things sit on the page. A plain repositioning pass built on CM dollars usually lifts category contribution 5-12% in the first quarter (Cornell School of Hotel Administration research). Go look at your menu tonight.

Sources: Toast, TouchBistro, meez, Cornell Hotel School research on menu engineering, Restaurant365.