Delivery vs dine-in menu: the same price myth and what the till says

How much margin does charging the same price on delivery and dine-in destroy?

Charging the same on both menus costs you between 9 and 18 margin points per order, and the 92-seat Italian place that opened this analysis proves it without rhetoric:

the same pappardelle at 18.90 USD left 11.40 USD of contribution margin at the table and 4.10 USD on the app, a 7.30 USD drop nobody watched because revenue was climbing 24%. The marketplace commission takes 15% to 30% of the ticket depending on the plan, packaging adds 0.45 to 1.80 USD per dish, disposable cutlery and the side sauce pile on cents that never reach the recipe cost, and transport shrinkage —the dish arriving cold, the container tipping over— closes the gap. As of August 2026 that arithmetic allows no nuance: delivery is a different business. The differential that works lives between 18% and 27% over the dine-in price, and it pays to break it into tiers because each one buys something different.

What each price-differential range buys you: from the useless 8% to the 27% that pays?

A markup of 8% to 12% —the one almost everyone applies out of fear— covers only basic packaging at 0.45 to 0.80 USD and never touches the commission;

with a 15% marketplace you still lose 3 to 5 margin points. The 15% to 20% band neutralizes a 15%-18% commission plus standard packaging, and it is the defensible minimum for operations with their own brand and shared logistics. From 21% to 27% you absorb commissions of 25%-30%, premium packaging of 1.20 to 1.80 USD with heat sealing, and the loss of the beverage mix. Above 30% the customer compares against the dining-room menu and you lose conversion before you gain the point. Remember that the percentage runs on final selling price, not on raw material cost. Commission is charged on the final price —tax included in most markets— so every dollar you add to pay for it pays commission again, and that loop is what ruins timid differentials.

The compounding effect: why raising 27 cents does not recover 27 cents

Take a 20 USD dish with a 27% commission: the platform keeps 5.40 USD. If you raise only those 5.40 to reach 25.40 USD, the commission climbs to 6.86 USD and you are still 1.46 USD short of the dine-in margin. The clean formula divides the dine-in price by one minus the commission rate, that is 20 divided by 0.73, which yields 27.40 USD. Only there do you break even, and packaging still has to be loaded on top. That is why 8% or 10% surcharges fix nothing: mathematically they cannot, even if the owner sleeps well believing he compensated. In the dining room, drinks and dessert account for 22% to 31% of the ticket at a food cost of 12% to 18%, and those two lines —not the signature dish— are what hold up the margin of the entire restaurant.

The sales mix warps and the main course is left alone

On delivery, a drink shows up in one out of five orders and dessert in fewer than one out of ten, because nobody orders flan on impulse after paying 4 USD in delivery fees and waiting 38 minutes. Add the tip, which in the dining room subsidizes part of the floor payroll and on the app goes to the courier. What you get is a ticket where the main course carries the whole fixed structure: rent, energy, hot line, supervision. If your delivery menu mirrors dine-in dish by dish, you are exporting the product with the lowest relative margin and keeping the two highest-margin ones at home. Five variables explain almost the entire gap between profitable delivery and delivery that funds the platform. The negotiated commission rate weighs 15 to 30 points and is the biggest one, with spreads of up to 12 points between the basic plan and the maximum-visibility plan.

The five factors that move your delivery price, with estimated impact

Packaging shifts 2 to 6 margin points depending on whether you use generic polypropylene or a leak-proof container with film. Order density —how many dishes per bag— changes the unit packaging cost by up to 40%, since bag, label and sealing get spread across items. Delivery radius alters shrinkage and returns. And input inflation rules: according to the USDA Economic Research Service, food-away-from-home prices rose 4.1% in 2024, while ACODRES reported 9.8% increases on Colombian restaurant dishes since February 2025. Negotiate with volume data, not with complaints. Platforms move commission in brackets and the argument that works is consistency: steady weekly orders, cancellation rate below 2%, stable preparation time. Ask for the reduced-commission plan in exchange for partial exclusivity or a 95% menu-availability commitment, and calculate beforehand what that exclusivity costs you in your own channel.

How to negotiate the commission and optimize the menu without fighting the customer?

In parallel, redesign the menu: pull the dishes with food cost above 32% that also travel badly, create delivery formats —the tray for two, the combo with a bottled drink— that recover the lost ticket, and set the differentiated price without apologizing.

The customer who compares already knows delivery costs money. Start today with your ten best sellers and calculate the real margin of each one across both channels. At Masterestaurant we treat dining room and delivery as two businesses with separate books, each with its own recipe costing, its own break-even and its own price. Diego F. Parra repeats it in every review of restaurant financial structure: if the channel changes the cost per plate served, the price changes, and if the price cannot change, the dish changes. The usual objection —«the customer will get upset»— confuses two different things, because the customer opening the app already agreed to pay 3 to 6 USD in delivery and a service fee; a dish priced 22% higher inside that context breaks nothing, while selling below cost does break payroll.

Two menus, two sets of books: the Masterestaurant position

I got this wrong for years, recommending price parity for brand coherence. Coherence does not pay suppliers. If your commission goes from 25% to 30% and you leave the price untouched, a dish that left 4.10 USD of margin drops to 3.10 USD, a 24% fall from a single contract line you never signed again. At 900 monthly orders that is 900 USD of cash gone without a single lost sale, and because revenue does not move, the dashboard stays green while the bank account empties. The restaurant already running a differentiated menu absorbs the hit by raising three points and survives; the one charging parity slides into negative margin and discovers that every new order costs it money. That is the stress test separating a delivery channel from a leak. Run the scenario with your own numbers before somebody runs it for you. Commission is charged on the final price, VAT included in most markets, so every dollar you add to cover it pays commission again.

Where marginal profitability per dish actually breaks?

Recovering 27 cents of commission on a 20 USD dish takes more than a 27-cent increase: you need 37 cents to land on the same absolute margin.

That compounding effect is why timid differentials of 8% or 10% fix nothing. Sales mix gets deformed. At the table, drink and dessert account for 22% to 31% of the ticket at a food cost of 12% to 18%, and they are what hold up the margin of the whole restaurant. On delivery the drink shows up in one order out of five and dessert in fewer than one in ten, so the main dish has to carry the entire structure alone. The standard recipe silently changes cost: the sauce portion goes up 15%-20% so the dish does not arrive dry, an extra container is added, and the crisp garnish is swapped for one that holds. Three invisible tweaks worth 0.60 to 1.10 USD, none of them in the original costing sheet.

Where marginal profitability per dish actually breaks — in practice?

Restaurant menu design for a screen is not menu design for print: on the app the customer sees a thumbnail, a name and a price, with no long description and no server suggesting anything.

Dishes that drain profitability sell better on screen because they tend to be cheap and familiar, while the stars of menu engineering need explaining and vanish from the cart.