Restaurant value proposition: before vs after running it through the till

The Medellín steakhouse: 268 million billed, 11 million lost every month

That steakhouse's declarative promise —«21-day dry-aged beef at neighborhood prices»— cost 11 million pesos a month on 268 million in billing, and the fault was not the cook's but the arithmetic nobody ran before printing the menu. The star dish carried 41% of sales and dragged a 38% food cost, six points above the 32% ceiling Masterestaurant sets as an absolute MAXIMUM, because the price had been anchored against a competitor buying beef with no aging at all and therefore playing a different cost game. In the costed version of that same promise, the aging stayed, portion size dropped from 320 to 280 grams and the price rose 9%; food cost closed at 30,4% and operating margin moved four points in eleven weeks. Same sentence, same beef, different arithmetic. Writing the promise before costing it locks in a food cost you can no longer escape without betraying the guest, and that sequencing error is the one that bleeds most money in this trade.

Sequence: calculate first, write second

When you announce «generous portions» without weighing the grams first, you have just signed a contract with your diner: cutting 40 grams six months later is the fastest way to lose a regular who had walked in every Thursday for three years. The costed route flips the order. You calculate contribution margin line by line, look at where the structure holds, and only then write the sentence that goes on the menu and the storefront. With U.S. menu inflation at +3,5% year over year in May 2025 —the slowest pace in sixteen months, per the National Restaurant Association— the room to fix things through price has narrowed: there is no cushion left to repair what was promised without math. A ceviche running 34% food cost that leaves 21.000 pesos per plate beats a pasta at 22% leaving 9.400, and yet most menus are sorted backwards, rewarding the pretty percentage over the cash that actually reaches the till.

Food cost percentage versus contribution margin in cash

The declarative proposition reasons in percentages because percentages are what the software shows and what people repeat in hallway conversations. The costed one reasons in pesos per unit sold, multiplied by real turnover. Run it at 60 plates a day: the ceviche contributes 1.260.000 pesos daily; the pasta, 564.000. Across 340 operating days the annual gap clears 236 million. That is the money lost to managing a menu with the wrong metric, and in Colombia, where Acodrés measured 59% informality in the sector during 2025, that cushion is the line between going formal and never being able to. Every value proposition carries a labor invoice, and the declarative one almost never acknowledges it. Promising «chef-driven cooking, every plate finished in front of you» means one more cook per shift, and that cook does not show up in the recipe's food cost: he shows up in prime cost, raw material plus productive payroll, which in a mid-ticket venue should land between 58% and 63% of sales.

Prime cost: a promise that demands hands also demands payroll

I have seen beautiful menus with 29% food cost and 41% payroll adding to 70% prime cost, meaning a business that already lost before paying rent. The costed version crosses both lines before signing the promise. If the operation cannot sustain the man-minutes the sentence demands, you change the sentence or you change the process; what you cannot do is leave the promise untouched and wait for payroll to behave. A promise that pushes break-even above the occupancy you actually reach on a Tuesday is a promise that fails, however extraordinary it reads on the storefront. With average annual revenue per restaurant near 1,76 million dollars according to Toast's sample of 859 venues, there is no room to fund décor, tableware and extra staff through declarative experiments. The costed route calculates how many covers per service the promise needs to pay for itself. If «linen napkins and sourdough baked in house» lifts fixed costs by 4,2 million pesos a month and your average contribution per cover is 18.000 pesos, you need 233 additional covers monthly just to break even.

Break-even as the filter for any promise

Do you have them? If the honest answer is no, that promise is not ambition: it is a liability written in large type. The costed version wins, and it wins by a mile. The digital channel is where an uncosted promise breaks fastest, because commission arrives after the menu has already been written. U.S. consumers spend 88,50 dollars a month on takeout and delivery, per Escoffier's 2025 dining trends report, and Southeast Asia's online delivery market moved 45.100 million dollars in 2025 according to Statista: real volume, not a passing fashion. But a 27% commission on a plate already carrying 32% food cost leaves 41% for payroll, packaging, rent and profit. The costed proposition builds a different delivery menu —fewer references, grams tuned to the packaging, prices with the commission baked in— and shields the dining room promise from cannibalization. The declarative one uploads the same plates at the same price and finds the hole three months later, when the accountant asks why sales grew while the bank account shrank.

Diego F. Parra: the question that opens every diagnosis

What does it cost you to deliver what you promise? That is the first question Diego F. Parra asks on walking into a restaurant, and across twenty years in this trade he has found that almost nobody has the answer at hand, because the branding workshop and the accounting department live on different floors of the same building. The Masterestaurant method forces both documents onto one sheet: the sentence the guest reads and the unit cost that sentence triggers. I got this wrong for years, recommending brilliant repositionings without opening the recipe costing first, and the outcome was always the same —promises the operator could not hold through low season—. The correction was simple and hard: no value proposition gets approved without contribution margin per plate, prime cost and break-even sitting beside it. That is not bureaucracy. It is the difference between a brand and a resignation letter.

What to choose for your operating profile?

If you run one venue, a mid-range ticket and under 90 days of cash, go straight to the costed proposition:

it is the one that moves operating margin between 4 and 9 points in a quarter, and the money to experiment simply is not there. If you operate three or more venues with a finance lead already tracking contribution by line, you are probably on the costed route already and your job is auditing that the storefront promise still matches this month's recipe costing, not the one from two years ago. There is only one case where the declarative version makes sense: a venture-funded opening buying market share at a deliberate loss, with a written cut-off date. Outside that, and with 74% of U.S. chain locations run by franchisees who do cost every line before signing it (Restroworks), writing without calculating means competing unarmed. Open the recipe costing of your best-selling plate and measure it this week.

Four differences that decide the outcome

The first difference is sequence, and almost nobody respects it: the declarative proposition is written first and funded later, while the costed one is calculated first and written afterwards. That sounds like a process footnote. It is not. Promise «generous portions» before costing the gram weight and you have just locked in a food cost you can no longer bring down without betraying the guest, and shrinking portions after promising them is the fastest way to lose a regular. The second lives in the unit of measure. The average owner watches food cost percentage; the mature operator watches contribution margin in currency per plate. A ceviche at 34% food cost that leaves 21,000 pesos beats a pasta at 22% leaving 9,400, yet 68% of the menus I review are ranked backwards, punishing the dish that contributes most because «its food cost looks high». The third shows up in the bad month.

Four differences that decide the outcome — in practice

A declarative proposition has no reverse gear: if your promise is «best price around» and chicken jumps 19%, you are trapped, because your differentiation IS the price. The costed version ships with a hierarchy of sacrifices defined in cold blood, when nobody was panicking, and that hierarchy is worth more than any tagline. The fourth is the one capital sees. According to Aaron Allen, founder of Aaron Allen & Associates, most restaurant concepts fail when scaling not for lack of demand but because unit economics were never sound in the original location, and multiplying a site that loses 3% only produces bigger losses. A restaurant investor reads exactly that: your proposition is not the point, whether the margin survives copy number twelve is.