How to make a restaurant profitable: the six mistakes that eat your margin and the method that lifts it

Step 1: calculate contribution margin in currency per dish, not the percentage

Your first deliverable is a new column on your menu: contribution margin per dish, meaning selling price minus the raw ingredient cost of that recipe. A dish running 44% food cost sold at 38,000 pesos leaves 21,280; another at 28% sold at 22,000 leaves 15,840, and the percentage deceives you because the second looks healthier while the first pays more rent every night. That is why control tracks two numbers together: per-dish food cost capped at 32%, a ceiling rather than a target, and the actual money that lands in the register per unit sold. Same-day verification: export four weeks of POS sales, cross them against your recipe cards, and if more than 15% of your dishes lack a card with current gram weights and costs, that is your real problem, not marketing. Multiply margin per unit by units sold over four weeks and rank descending: those top four dishes are the only ones your video content should push for the next quarter.

Step 2: rank the menu by total margin and keep only the top four

Almost nobody applies this filter before deciding what gets filmed, and its absence explains why well-executed campaigns end up subtracting operating profit. The arithmetic is merciless, which helps: if dish number one leaves 21,000 pesos per unit and dish number twelve leaves 6,400, shifting 200 monthly units from the second to the first is nearly three million pesos of clean money requiring not one additional customer. Verifiable deliverable: a twelve-row sheet, sorted, with the cutoff date at the top and the four dishes marked in CAPITALS. If it does not fit on one screen, you overcomplicated it. Prime cost, which adds food plus fully loaded payroll, must land under 60% of net sales, and that number gets calculated BEFORE you approve any content budget. The reasoning is elementary arithmetic: the sector runs on single-digit net margins, according to Hudson Riehle, senior vice president of research at the National Restaurant Association, so a prime cost sitting at 66% leaves so little cushion that no additional volume rescues it.

Step 3: lock prime cost below 60% before spending a peso on production

Cost pressure is real too: the U.S. producer price index for all foods sits 35% above its February 2020 level (USDA ERS / BLS 2026), while in Colombia restaurant dishes rose 9.8% in February 2025 according to Acodrés. Today's control: divide last closed month's food plus payroll by net sales; if it exceeds 62%, freeze video production and fix the kitchen first. Six of every ten pieces this month go to your four highest-margin dishes, and the remaining four split brand, team and seasonal work. Production quality matters far less than anyone selling it claims: a Reel shot on a phone with window light converts as well or better than a 900-dollar production, provided it shows the right dish and the plating looks like what actually leaves the kitchen. The lens was never what broke the chain. Deliverable: a twelve-piece calendar with the dish name on each row and its margin beside it, because seeing that number next to the script changes what the person filming decides.

Step 4: assign 60% of your Reels calendar to those four dishes and shoot them on a phone

And one hard rule from the Masterestaurant framework Diego F. Parra applies in every engagement: no dish enters the calendar if its recipe card is more than ninety days stale. Your closing indicator is neither reach nor saves: it is the share of orders your four highest-margin dishes represent, compared against the baseline you captured in step two. Reach and profile visits live on one dashboard, contribution margin lives on another, and that disconnect costs more money than any expensive supplier. If those four moved from 24% to 33% of orders while total sales stayed flat, you already won: the same register books more profit without a single new customer. Verify with the POS mix report every Monday, on the same sheet, using the same cutoff date. One field detail people skip: delivery distorts the mix because over 40% of adults order takeout three to five times a month (UpMenu 2024), so separate your channels before reading the number.

Four mistakes that wreck this guide and how to avoid them

The costliest mistake is celebrating sales growth without checking operating profit, which is exactly what happened at that pasta restaurant in Medellín: 22% more revenue across four weeks, three points less profit, and an owner blaming his kitchen payroll. The second mistake is repricing the viral dish instead of changing which dish gets promoted, because raising a proven-demand product 15% burns trust and never fixes the recipe. Third, promoting a high-margin dish the kitchen cannot sustain at volume: if ticket time goes from seven to fourteen minutes, the table that stops turning eats the margin. Fourth, measuring at seven days. Menu mix needs four weeks to settle, and any reading before that is noise dressed up as data. Suppose you fix the content, push the four correct dishes and touch nothing else operationally: the mix improves, a few margin points arrive, and in month two your protein supplier raises prices 6% and you land back where you started without understanding why.

The counterfactual that sets priorities: what happens if you only change the video

That is the ceiling of working a single dashboard. The apparent tension between marketing and costs resolves through sequence rather than balance: dish arithmetic first, script second, never the reverse, because effective video selling a badly costed dish accelerates the loss instead of braking it. With AI-assisted scheduling reporting labor cost reductions of 8 to 12% and forecast accuracy above 90% (TimeForge 2025), the payroll lever exists and it is real, but it arrives later. Sort out dish margin first. You finished this guide properly when you can answer five things with a number, without opening another folder: what the per-unit margin of your four main dishes is, what each one's food cost is and that none exceeds 32%, what your prime cost was last closed month and that it stayed under 60%, what share of this month's content calendar went to those four dishes and that it reached 60%, and how your menu mix moved between baseline and the four-week cutoff.

Closing checklist: how to know everything landed

If a single one fails, the chain is open and results will stay inconsistent. Your concrete action for today: open the product-level sales report for the last four weeks, add the margin column and sort it. That sort, forty minutes of work, decides what you film next quarter. Production quality is not the differentiator, and that surprises most owners I work with. A Reel shot on an iPhone 13 with window light converts as well as a 900 USD production, provided it features the right dish. What breaks the chain is SELECTION: nobody filters the catalog by margin before deciding what to film, so digital marketing performs flawlessly while promoting a product that loses money on every unit. The second break is attribution. According to Hudson Riehle, senior vice president of research at the National Restaurant Association, the industry runs on net margins that rarely reach double digits, which means menu-mix decisions move profit more than almost any other lever available.

Where the chain between video and register actually breaks?

Without a redemption code or a campaign-specific POS item, you cannot tell whether those 400,000 views brought 60 tables or 6, and that blind spot turns every content budget into guesswork.

A third fracture is less comfortable to name: the owner producing the content is usually the same person cooking, and that hour is worth far more on the line at 8 p.m. on a Friday. Filming four pieces in a two-hour Tuesday morning block is not an aesthetic preference, it is opportunity-cost arithmetic. An improvised shoot during service costs between 180 and 340 USD in lost sales through slower table turns. Then comes the fracture that moves the most money. Content changes the MIX of what sells, and mix is the fastest profitability lever a restaurant has, faster than renegotiating with suppliers and far faster than raising prices. Shift three points of share from a 41% food cost plate to a 27% one and, in a location doing 45,000 USD monthly, that is roughly 1,900 USD of additional margin per month without selling a single extra plate.