Contribution margin per dish: the traditional method sees food cost, the Masterestaurant method sees the cash

1. What governs cash: food cost or contribution margin per dish?

Cash is governed by contribution margin per dish, not by food cost taken alone. Contribution is how much money each dish leaves to cover rent, payroll and profit once only the variable cost of its ingredients comes out.

A dish at 22% food cost can leave as little as $4; another at 34% can leave $19 — running the menu by that percentage promotes, almost every time, the wrong plate. That's where the ratio lies; the dollar doesn't. I tell every board I visit the same thing: the number looks pristine on the spreadsheet while the cash drains out the back. The sector averages barely 10.66% pre-tax operating margin, per NYU Stern (Damodaran, 2024 dataset), and with a cushion that thin, every contribution dollar lost to deciding backwards is what separates closing the month in the black from closing it in the red. Multiplying margin by real rotation finishes what the ratio only half tells, and that's where the dollar actually reaching the drawer shows up.

2. Contribution without turnover is half the equation

A dish leaving $19 but sold just 8 times a day adds $152 daily; one leaving only $6, turned 60 times, adds up to $360 — unit-margin vanity loses to volume. I cross contribution margin against each dish's last 90 days of real rotation to rank the menu by total dollars, never by whichever ratio shines brightest. If you look only at the percentage, the most common reading error is falling for the highest-margin dish without noticing almost nobody orders it. That error costs real money: 90% of full-service operators raised prices in 2024, and 60% pulled dishes from the menu (National Restaurant Association, 2024). Cutting by percentage removes, almost always, the wrong low-rotation dish and sacrifices total contribution that was already secured. Payroll stays off the dish because, spread there, it distorts every read of profitability; its place is the break-even. The traditional method divides labor cost across every recipe and makes strong contributors look expensive.

3. Why payroll is not charged to the dish

We isolate only the variable ingredient cost to decide the menu, and send payroll, rent and utilities to the calculation of how many covers are needed not to lose money. This isn't bookkeeping trivia. Payroll already weighs more than 25% of expenses in 2024, up from 23% in 2021 (Toast, via Restaurant Dive), and 98% of operators reported labor costs rose (National Restaurant Association, 2024). At that weight, charging labor to the dish muddies every menu decision. Profitable operators run payroll at 34.2% of sales against 36.5% for the average (National Restaurant Association, 2024 data): 2.3 points won by managing the break-even, not by splitting the cost recipe by recipe. Dine-in and delivery don't share the same contribution, and that's what decides a menu's real profitability today. With roughly 75% of traffic happening off-premise (Circana), a dish that leaves $12 in the dining room can turn into an app-side leak the moment the platform commission, packaging and in-transit loss come out.

4. Channel changes contribution: dine-in is not delivery

We calculate contribution per channel — dine-in, delivery and take-away with separate accounting — so no dish travels at a loss without the owner knowing. This isn't a hunch: the global ghost-kitchens market hit $72.06 billion in 2024, per Credence Research, proof of how much weight off-premise consumption has gained. The same dish can be the star of the table and, at once, the app's quiet bleed. Without contribution margin by channel, the restaurant ends up subsidizing the platform with its own menu. Treating food cost as the sole KPI optimizes a ratio, not dollars, and that's why it became an obsolete decision architecture. Chasing a low percentage pushes owners toward blind price hikes or portion cuts, when the real goal is maximizing the total contribution reaching cash each service. At Masterestaurant food cost stays a ceiling — a maximum of 32% per dish — never a target; the menu is governed by contribution margin times turnover.

5. Food cost as the only KPI is an obsolete decision architecture

Failing to change the frame isn't paid in an ugly ratio inside Excel: it's paid in dollars leaking every shift. That's the key to why a restaurant's average sale multiple hovers around 2.80x–3.65x EBITDA, per Sofer Advisors: buyers don't pay for food cost, they pay for the sustained contribution the menu generates. A well-run fast-casual sells at 4x–7x EBITDA; the one steered by percentage leaves that multiple on the table. When contribution governs the menu, the board stops arguing food cost and starts deciding on dollars. Menu engineering reorders itself from there: the high-contribution, high-rotation dish gets protected, the low-contribution one gets redesigned or retired regardless of its percentage, and each channel gets evaluated on its own. This is exactly what I bring to Masterestaurant boards in a three-minute read, because the owner already knows food cost yet keeps deciding by ratio.

6. What changes in the boardroom when contribution governs the menu?

Context gives no room: 99% of operators spent more on labor during 2024 (TouchBistro), and the sector's operating margin barely touches 10.66% (NYU Stern, 2024).

With that cushion, the one lever left standing is total contribution. Today's task isn't theoretical: reorder the menu by contribution margin times rotation, and separate contribution by channel before the next menu change. Percentage deceives; the dollar doesn't. One dish at 22% food cost can leave as little as $4, another at 34% can leave $19 — chase the ratio and you'll almost always end up pushing the wrong plate. Rotation finishes what margin only half tells: I multiply contribution per dish against the last 90 days of real turnover, and that's where the dollar that actually reaches the drawer shows up, not the one the ratio promises. I never load payroll onto the dish. The traditional method spreads it and muddies the read; sending it to break-even, where it belongs, leaves the dish clean to decide on.

7. The differences that move EBITDA

With roughly 75% of traffic off-premise (Circana), a dish that performs in the dining room can turn into a leak the moment the delivery commission comes out — which is why we calculate contribution channel by channel, never in one lump. AI-assisted menu engineering swaps the static sheet for a shortlist of decisions: what to reprice, what to redesign, what to retire. That's the real difference — decision architecture, not another report to file away.