How to increase restaurant sales on Rappi: the pricing mistakes that eat your margin and the right method

What does it really cost to sell a dish on Rappi?

A dish listed at 35,000 pesos on Rappi leaves between 9,000 and 12,000 pesos of contribution when the digital price was set properly, and between 1,000 and 3,000 when the dine-in price was simply copied over.

The arithmetic is boring, which is precisely why almost nobody runs it: out of those 35,000 goes the aggregator commission, which in Colombia sits in a band of 22% to 30% depending on the plan and on whether the app supplies the courier; out goes packaging, which as of September 2026 runs between 900 and 2,200 pesos per order depending on whether you use sealed kraft board or generic plastic; out goes food cost, which in a properly costed menu never exceeds 32% of the selling price. Add it up: 27% plus 5% packaging plus 32% food cost is 64 points before payroll, rent or utilities.

What does it really cost to sell a dish on Rappi — in practice?

Whatever remains has to pay for the kitchen that produced that order. There are three digital markup ranges and each one buys something different.

The first, between 0% and 8% over the dine-in price, only works if you negotiated commission below 18% or you run your own delivery fleet; at standard commission that range funds the platform with your margin. The second, between 15% and 25%, is what I recommend by default: it recovers commission, pays for packaging and leaves dine-in contribution untouched, which in practice means a 30,000-peso table dish gets published between 34,500 and 37,500. The third, above 30%, makes sense in dark kitchens with no reference dining room, where nobody compares prices, or in very high gross margin categories such as drinks and desserts, where alcohol was named a top-margin category by 46% of surveyed operators (Technomic / Nation's Restaurant News, 2024).

What each digital price range buys you?

Outside those cases, going past 30% punishes conversion without buying the business anything. The right digital price is not a universal number, and FIVE variables push it up or down with measurable impact.

Negotiated commission rules: every percentage point you shave off gives you back nearly a point of markup, and moving from 30% to 24% lets you publish 6% cheaper without losing a peso. Packaging weighs between 3% and 7% of the ticket, and there the mix matters, because a soup needs a sealed container and a burger does not. Average ticket dilutes everything: an 80,000-peso order absorbs commission far better than a 22,000-peso one. Active promotions add another 8% to 15% when the owner accepts a 2-for-1 from a pop-up without recalculating. And Colombia's 8% consumption tax, if the restaurant absorbs it instead of passing it through, eats the rest.

The five factors that move your price inside the app

At Masterestaurant we cost those five before touching the menu. A steakhouse on Calle 85 was billing 41 million pesos a month on Rappi and losing money on every single order, and the breakdown explains why nobody caught it for seven months. The app settles weekly and the owner watched the deposit, not the composition: 27% commission, another 12% in promotions he had activated himself from a pop-up in the console, and 1,800 pesos of packaging that never entered the costing sheet. The flagship dish left 4,100 pesos of margin at the table and 380 in the application, that is, 9% of what it made thirty meters away. After raising the digital menu 22%, switching off both permanent promotions and cutting the list from 46 dishes to 19, orders fell from 1,480 to 1,010 a month and channel contribution went from negative to 6.4 million.

The Bogotá steakhouse case: 41 million and a loss

Fewer orders, more cash: that is the equation nobody wants to hear. Raising the digital price does reduce conversion inside the app, that is measured and true, but the order you lose to a price 20% higher was exactly the order that made no money. Here sits the tension almost no consultant will name: the aggregator rewards volume with placement in the listing, and the owner reads that placement as business health when he is really buying visibility with his own margin. I would rather have 700 profitable orders than 1,500 that bankroll somebody else's operation. The bridge between both ideas is the menu, not the price alone: cut down to the dishes that travel well, the ones that arrive at minute fifteen the way they left, then lift the ticket with sides and drinks, and the conversion drop gets offset by an average ticket 18% to 25% higher.

Why losing orders when you raise prices is the right signal?

That is the order. Price, menu, ticket, and only then volume. Commission is negotiable and most owners never try, because they assume the contract is a form.

It is not: aggregators run internal bands and the account executive can move 3 to 6 points if you show up with three things on the table. First, documented volume for the last six months, with orders, average ticket and cancellation rate on your side; an operation cancelling below 3% is worth money to the platform. Second, a real alternative: if you have your own WhatsApp or web channel carrying 20% of your orders, say so. Third, willingness to join platform campaigns with a capped budget, never an open one. And one rule that is not up for negotiation: never accept a promotion from a pop-up before calculating contribution with that discount applied. That button has cost more margin than every commission combined.

The minimum dashboard: four numbers and a weekly review

Holding Rappi sales without bleeding cash takes FOUR numbers reviewed every Monday, and none of them is the one the app puts front and center. Contribution per order comes first, in pesos rather than percentage: published price minus commission, packaging, food cost and tax. Second is the digital channel's average ticket set against the dining room's; when digital sits lower, the menu was built wrong. Third is the share of orders carrying an active promotion, which in healthy operations stays under 15%. Fourth is your rating, since each additional star is associated with a revenue increase of 5% to 9% (Harvard Business School, Michael Luca, Reviews, Reputation, and Revenue). Diego F. Parra puts it this way in foodtech diagnostics: if a channel cannot answer those four numbers in two minutes, that channel is not being managed, it is being endured. Start with the costing file, not with the app console, and work in the order that follows.

What I would do this week with your digital menu

Take your ten best-selling dishes in the digital channel and calculate the real contribution of each one with commission, packaging and tax included, as of September 2026 and with this month's ingredient prices, because Colombian restaurant dishes rose 9.8% since February 2025 according to ACODRES and last year's costing describes nothing anymore. Anything leaving less than 25% contribution comes off the digital menu or goes up in price. Then publish the markup between 15% and 25%, switch off every promotion without an end date, and wait three full weeks before judging the outcome: week one always dips. If by week four total channel contribution has not risen, the problem was never the price but the mix, and there we go back to the menu. Commission is not the problem; the sequence of decisions is. A virtual restaurant that prices its digital menu properly absorbs 27% comfortably, because the channel brings demand its street never will.

Where the math actually breaks?

The one that prices badly and then chases volume enters a spiral: more orders, more commission, more kitchen hours, less cash. There is a real tension here, and it deserves naming instead of dodging.

Raising the digital price does cut in-app conversion — measured, documented — but the order you lose to a 20% higher price was precisely the order that made no money. I would take 700 profitable orders over 1,500 that bankroll somebody else's operation. The bridge between both ideas is the menu: trim to dishes that travel well and lift average ticket with combos, and conversion comes back without cutting price. The third point almost everyone ignores: the aggregator owns the customer, you do not. So the goal cannot be living off Rappi, but using it as an acquisition channel while you build direct ordering. A dark kitchen at 60% aggregator and 40% owned channel has a business; one at 95% aggregator has a subcontracting agreement with the platform.

Where the math actually breaks — in practice?

If digital menus come up: ALWAYS keep the physical menu in the dining room alongside the QR. The printed menu controls the guest experience — service pace, menu narrative, suggestive selling;

the QR is a complement for delivery, price updates and analytics. Never QR only.