From −2% to 11% EBITDA: how to price a tasting menu that stopped bleeding cash, with the Restaurant Model Canvas and the Standard Recipe Generator

The diagnosis: 43% food cost where the house believed 36%

A tasting menu's price is set from the real food cost of the full journey, not dish by dish: costing pass by pass understates the cost by 5 to 9 percentage points. This fine dining room of 18 tables (44 covers), 11 employees and six years of operation believed it ran a 36% food cost; auditing standard recipes put it at 43%. The house filled the room four nights a week with 4.8 reviews, yet the owner topped up cash from his own pocket every month-end. The full-service sector benchmark is 32.0% of sales (National Restaurant Association, 2024), and a seven-course tasting multiplies the waste that average never contemplates. The question the owner brought —should I raise the price?— was the wrong one. The right one: what does it truly cost to serve this seven-pass journey down to the last petit four. A tasting menu is not the sum of seven dishes, it is a single experience with compound waste, and that was the leak.

Why costing pass by pass drained the till?

The kitchen quoted each pass by its theoretical food cost and ignored three costs that only surface across the full journey:

perishable mise en place waste (herbs, foams, gels prepped for 44 covers and binned if unsold), precision-plating waste (the portion discarded for missing the visual standard) and the un-costed pairing. Food waste costs the U.S. restaurant industry roughly $162 billion a year (The Restaurant HQ, 2025); in a tasting that share concentrates because every course demands perfection. Adding pass by pass, the house saw 36%. Adding the journey with its compound waste, the real number was 43% (per the case audit). Those 7 points were exactly what the owner replaced each month. The intervention started with the Masterestaurant standard-recipe tool, applied to the full journey rather than each isolated dish. With Diego F. Parra we costed the seven passes as one unit: every recipe with real yield, mise en place waste assigned per cover, and plating discard charged to the course cost, not hidden under 'miscellaneous'.

The action: re-costing the journey with the Masterestaurant tool

The pairing was costed as its own line with a target margin. With the whole journey on the table, we anchored the ticket to a target prime cost of 58-62% —food cost plus labor— read from audited recipes, not from the neighbor's price. Full-service labor cost runs around 36.5% of sales (National Restaurant Association, 2024), so with food cost controlled the prime cost closed within a healthy range. The ticket rose from $95 to $128, but the real work was the costing, not the final number. A well-costed pairing is where fine dining recovers the margin the precision plate takes away, and here it was the giveaway link. The house included three 'courtesy' glasses whose bottle never entered the costing; the liquid's food cost lived off the recipe card. We re-costed each pairing with its real wine cost, service waste and a target margin, turning it into a sellable line with two tiers: standard pairing and reserve.

The pairing: from courtesy to margin center

Healthy occupancy cost should not exceed 6-10% of gross sales (Toast benchmarks), and only with the entire cost structure visible could we verify the new price covered rent, labor and beverage without cannibalizing margin. Giving the drink away or charging it at cost means forfeiting the format's profit engine: the pairing went on to contribute 22% of the per-table sale (per the case), margin that used to evaporate inside the wine list. The result was journey food cost from 43% to 31% and prime cost within the 58-62% target in three months (per the case). With the ticket anchored at $128 and the pairing sold as a line, the till stopped needing the owner's injection: operating margin went from negative to positive without losing a single cover of the four full nights. The room kept its advance reservations and reviews held at 4.8, because the change never touched the chef's cooking, it touched the costing.

The measurable result after 90 days

To size the business: opening a fine dining room of this caliber costs, in the top quartile, roughly $750,500 ($177 per sq ft) in the U.S. (Rezku, 2025); recovering 12 food-cost points protects a good share of that investment every year. The gap between replacing cash and earning it came down to one calculation: cost the journey, not the dish. The transferable lesson is one: cost the tasting menu as a full journey with compound waste, and anchor the price to your target prime cost (58-62%), not to the neighbor. Small independent (one room, chef-owner): this week audit the standard recipes of your three costliest passes and include mise en place waste and plating discard; the real food cost is almost always 5-9 points above the theoretical. Mid-size (11+ employees like the case): this week build a pairing line costed with its own margin and stop giving the drink away; that is where the margin the precision plate takes lives.

Transferable lessons by operation size

Multi-site group: this week standardize journey costing in a single template per location and compare real food cost across sites —the dispersion tells you where the till leaks. Replacing an employee costs US$2,305 in hard costs (Black Box Intelligence, 2024), so stabilizing margin also stabilizes the team. This result does not replicate identically in every context, and saying so avoids survivorship bias. First: I would not expect it in a restaurant already near the sector's 32% food-cost benchmark (National Restaurant Association, 2024) with an already-costed pairing —there the journey lever yields little because the leak is already closed. Second: it does not apply cleanly to high-volume, low-ticket formats (limited service, median food cost 32.4%; National Restaurant Association, 2024), where there is no precision-plating waste nor pairing to recover; the journey-costing model loses meaning. Third: in markets where the diner does not pay for pairing —bring-your-own-wine culture or free corkage— the margin center that saved this case simply does not exist, and raising the ticket without that line can empty the room.

Limits of this case

The method is sound; its performance depends on there being compound waste and sellable beverage to recover. Unit costing vs. course costing: a tasting menu is not the sum of seven dishes, it's a single experience with compound waste. Costing course by course underestimates real food cost by 5 to 9 percentage points — exactly the gap that emptied this case's cash. The pairing as a margin center, not a courtesy: the well-costed beverage is where fine dining recovers the margin the precision dish takes away. Giving it away or charging it at cost forfeits the format's profitability engine. A price anchored to target prime cost vs. to the neighbor: pricing by watching the competitor ignores your real cost structure. The right anchor is your target prime cost (58-62%) read from audited standard recipes.