Rappi delivery strategy: what actually moves margin in 2026
A Rappi delivery strategy pays off when you cost the channel SEPARATELY from the dining room: 18% to 30% commission on menu price, $0.20 to $0.60 of packaging per order, and a target food cost of 26% to 28% on delivery items against the 30-32% the dining room tolerates. Copy the dine-in menu at dine-in prices and every order shaves utility even as sales climb; with a channel menu, channel pricing and at least 40% of orders coming from pickup or your own channel, the app stops being a tax and becomes paid customer acquisition.
A grill house in Bogotá was billing 41 million pesos a month through Rappi and losing 3.8 million in that same channel. The owner found out only when we split the P&L by channel, because the consolidated number hid the hole: dine-in was subsidizing delivery and nobody had decided that.
Selling on Rappi was never the mistake. Selling on Rappi at dine-in prices, with the same 74-item menu, ignoring packaging and reading food cost against menu price instead of the NET the platform settles — that is the mistake. When you compare raw material cost to the price on the screen rather than the money that actually lands in your account, you are lying to yourself with a calculator.
The 2026 question is no longer whether to be on the apps. It is how much of your installed capacity you hand to a channel that never gives you the customer record, and at what channel price. Rappi operates across nine Latin American countries and iFood clears more than 60 million orders a month in Brazil: the volume is real, what matters is how much of it stays with you.
Here is the thesis, ahead of the arguments: a Rappi delivery strategy works as a COSTED CHANNEL with its own menu and its own price structure, and fails as an automatic extension of the dining room. Everything below is the arithmetic behind that sentence.
Side-by-side comparison
| Rappi delivery with no channel costing | A costed Rappi delivery strategy | |
|---|---|---|
| Effective commission on gross sales | ✕18% to 30% nobody subtracts before pricing | ✓18% to 30% loaded into channel price from day 1 |
| Real food cost per order | ✕38% to 44% against net settled revenue | ✓26% to 28% against net settled revenue |
| Items on the digital menu | ✕60 to 90 dishes copied from the dining room | ✓18 to 24 dishes that travel well and rotate |
| Packaging cost per order | ✕$0.20 to $0.60 absorbed by the restaurant | ✓$0.20 to $0.60 passed into price or charged apart |
| Contribution margin per order | ✕-4% to 6% depending on the day and active discount | ✓22% to 31%, steady and measured weekly |
| Channel dependence | ✕85% to 100% of delivery comes from the app | ✓55% to 60% app, the rest owned channel and pickup |
| Kitchen dispatch time | ✕14 to 22 minutes with the dine-in line saturated | ✓8 to 12 minutes with a dedicated assembly station |
| Customer data | ✕0 owned records after 12 months | ✓1,200 to 3,000 owned contacts a year via pickup and QR |
Channel pricing stops being heresy and becomes the norm
The trend moving the most money in 2026 is also the simplest: charge differently on the app than in the dining room, and do it openly. The measurable signal sits in the cost structure rather than in an opinion survey, because platform commissions run between 18% and 30% of the sale price, and when San Francisco capped those commissions at 15% back in 2020 (Restaurant Dive), it proved that percentage weighs enough for a city to legislate around it. A dish priced at $32,000 at the table, with a 26% retention and $1,400 in packaging, has to go out at $39,000 on Rappi to hold the same contribution margin in pesos. If you run a single location, raise your ten fastest-moving items first and watch the average check for two weeks. With three or more units, build a complete channel price list from day one. Curating the delivery menu is the second profitability lever and almost nobody pulls it in time.
The menu that travels gets shorter: fewer items, faster turns
What backs the decision is the check itself: Deliveroo closed 2024 with GTV per order of GBP 27.6, up 5% year over year (Deliveroo plc 2024), while the average United States order sits between USD 20 and USD 35 (Lightspeed 2025) and Spain hovers near USD 24 per online order (Ken Research 2025). That narrow band tells you something brutal, which is that the app customer buys a check, not a menu. An operation running 74 items on Rappi pays for kitchen complexity to sell the same check it would sell with 22. In Masterestaurant audits of delivery operations, the first decision is usually cutting from seventy-something down to fewer than thirty items, starting by killing everything that arrives soggy, dry or overcooked. Splitting the income statement by channel stopped being big-chain sophistication and is now a survival requirement.
P&L by channel: the accounting that separates the winners
A grill house in Bogotá billed 41 million pesos a month through Rappi and lost 3.8 million in that same channel; the owner found out when we opened the P&L by channel, not before, because in the consolidated view the revenue covered the hole and the dining room subsidized delivery without anyone having decided it. I got this wrong for years myself, reading food cost against menu price instead of against the NET price the platform settles. With a 26% commission and $1,400 in packaging, a $32,000 dish actually leaves you $22,280; the 30% food cost you thought you had is really 43%. Delivery tolerates a target food cost of 26% to 28%, against the 30-32% the dining room does absorb, and those four points are the entire strategy. The most expensive consequence of living on Rappi never shows up in the weekly settlement: you hand over installed capacity in exchange for volume you can never contact again.
The app keeps the customer data, and that carries a price
The global delivery services market went from USD 380.43 billion in 2024 toward a projected USD 618.36 billion by 2030, at a 9.0% CAGR (Grand View Research 2025), and prepared meal delivery already accounts for more than 64% of online delivery revenue (Grand View Research 2024). That growth is real and so is the temptation to ride it. But a restaurant doing 60% of its sales through an app with zero owned customer base is a platform supplier, not a brand. The practical move: drop a physical insert into every package with a direct-order incentive, then measure how many migrate within ninety days. Ghost kitchens consolidated as a cheap expansion format and the numbers are no longer a promise. In Brazil the home segment of dark kitchens moves USD 5,702 million, and in India it reaches USD 12,000 million, with another USD 4,500 million in the multi-brand segment (Global Growth Insights).
The dark kitchen matures and stops being a pandemic experiment
Launching a virtual brand is no longer a garage bet; it competes against operators with capital. The consultant's read is uncomfortable: a dark kitchen lowers your dining room rent and payroll while pushing channel dependency to practically 100%, so it multiplies the risk we just described. If your kitchen sits idle between 2 and 5 in the afternoon, a second virtual brand on that same equipment is the play with the best return. If your kitchen already runs at 85%, leave it alone. Costing packaging per order, rather than as a monthly overhead line, is the accounting change that fastest corrects the illusion of profitability. Between $900 and $2,400 per order, depending on whether the dish travels in compostable cardboard, in lidded aluminum or in a double container with a sauce divider, the annual gap across 1,800 monthly orders reaches 32 million pesos. That figure appears on no platform dashboard.
Packaging enters the costing or the costing does not exist
Run it backwards: if packaging jumped 40% tomorrow on imported resin costs, would you know exactly how many pesos your margin drops per item, or would you wait for month-end close to find out through the result? Anyone who cannot answer that in two minutes is not costing, they are estimating. Load packaging onto the recipe card of every delivery dish, with its own code and unit cost, exactly like protein. Adopt three things immediately: differentiated channel pricing, delivery recipe cards with packaging included, and a P&L split by channel. None requires investment, only a decision, and together they move between four and seven margin points. Watch without investing yet in artificial intelligence automation of delivery operations, which arrives preceded by enormous noise: global agrifoodtech investment hit its record of USD 51,000 million in 2021 and has been correcting ever since (AgFunder News), meaning many promised tools will never reach a mature product.
What to adopt now and what to merely watch through 2026?
According to Diego F. Parra, consultant at Masterestaurant, sequence matters more than the tool: first the channel arithmetic, then the software that automates it, because automating a wrong costing only produces wrong reports faster.
The United States QSR market moved US$ 289,680 million in 2024 (Business Research Insights) on that basic discipline, not on algorithms. Negotiating the commission percentage is the conversation that eats the most time and yields the least margin, and it deserves to be said without diplomacy. Rappi operates across nine Latin American countries and iFood exceeds 60 million monthly orders in Brazil; against that volume, an independent restaurant with 1,500 monthly orders holds no leverage, and the two or three points it occasionally wins evaporate in the next discount campaign the app suggests it finance. Europe's meal delivery segment billed close to US$ 49,000 million in 2024 (Statista 2024), and those platforms will not adjust their model for you.
The overrated trend: fighting the platform over commission
Commission is not negotiated with hope, it is passed into price. Stop chasing the meeting with the account manager and spend that week reworking your ten best sellers, with channel pricing and costed packaging. That is where the margin points are waiting. CHANNEL PRICE, not a single list. Commission is not negotiated with hope: it is passed through. If the platform withholds 26% and your packaging runs $0.35, a $8 dine-in dish needs to list near $9.75 in the app to hold the same contribution margin in cash. Delivery customers compare you with other delivery, never with your dining room, and according to Diego F. Parra, restaurant consultant at Masterestaurant, price resistance shows up far later than owners fear: it shows up past a 30% gap. A MENU THAT TRAVELS. Fries arrive soggy, risotto arrives dry, medium steak arrives medium-well. Costing the channel without curating the menu is makeup.
Four differences that decide whether the channel leaves money
In Masterestaurant delivery audits the first decision is usually cutting from 70 items to 20, and revenue does not fall: it concentrates. CAPACITY, not ambition. Every app order landing at peak competes with a table that leaves more margin. If your kitchen dispatches 34 plates an hour and the dining room is full at 8:30 p.m., taking delivery without a dedicated station costs you table turns. The fix is not switching the app off: it is switching it off from 8:00 to 9:30 p.m. and on from 2:00 to 5:00 p.m., when your kitchen sits idle. OWN THE DATA. I got this wrong for years: I treated the app as a necessary evil and left it there. It is not. It is paid acquisition. You pay 26% to meet a customer once, and your job is converting that customer to your own channel before the third order — with the printed menu inside the bag, a QR to direct ordering and a concrete reason to come back without a middleman.
Head to head: the decision in five criteria
The myth: Rappi sells for youWhat gets repeated
- «I upload the whole menu and let the market decide»: 74 items stretch dispatch times and sink your rating.
- «The platform pays for the 30% discount»: in most campaigns you co-fund between 50% and 100% of it.
- «Sell more and margin shows up»: with negative contribution margin, every extra order deepens the loss.
- «Rappi brings me customers»: it brings transactions; the customer belongs to the app until you capture the record.
- «Same price in app and dining room out of respect»: respect means charging what it truly costs to serve that channel.
What the channel measurably doesMasterestaurant
- Channel pricing runs 15% to 25% above dine-in and customers accept it: they pay for convenience, not for the dish.
- A digital menu of 18 to 24 items lifts ticket because it concentrates sales on what actually leaves margin.
- Packaging is raw material: $0.20 to $0.60 per order, and bowls with separate sauce push past $0.80.
- A virtual brand on the same kitchen adds revenue at near-zero fixed cost, provided it shares 70% of the inventory.
- Pickup and owned channels give back 18 to 26 margin points on the very same dishes.
Side-by-side comparison
| Rappi delivery with no channel costing | A costed Rappi delivery strategy | |
|---|---|---|
| Effective commission on gross sales | ✕18% to 30% nobody subtracts before pricing | ✓18% to 30% loaded into channel price from day 1 |
| Real food cost per order | ✕38% to 44% against net settled revenue | ✓26% to 28% against net settled revenue |
| Items on the digital menu | ✕60 to 90 dishes copied from the dining room | ✓18 to 24 dishes that travel well and rotate |
| Packaging cost per order | ✕$0.20 to $0.60 absorbed by the restaurant | ✓$0.20 to $0.60 passed into price or charged apart |
| Contribution margin per order | ✕-4% to 6% depending on the day and active discount | ✓22% to 31%, steady and measured weekly |
| Channel dependence | ✕85% to 100% of delivery comes from the app | ✓55% to 60% app, the rest owned channel and pickup |
| Kitchen dispatch time | ✕14 to 22 minutes with the dine-in line saturated | ✓8 to 12 minutes with a dedicated assembly station |
| Customer data | ✕0 owned records after 12 months | ✓1,200 to 3,000 owned contacts a year via pickup and QR |
The numbers you decide with, not the ones people repeat
“We closed 41 million pesos on Rappi and were losing 3.8 million right there. We split the P&L by channel, cut the digital menu from 74 items to 21, raised app prices 19% and set up an assembly station with a half-shift assistant. Three months later channel revenue dropped to 36 million, but contribution margin went from -9% to +24%, meaning 8.6 million that used to evaporate. And with the printed menu inside the bag we captured 1,470 customers into our own WhatsApp list.”
How to build the strategy in 90 days without killing revenue
Take three months of Rappi settlements and build a separate column: gross sales, withheld commission, co-funded discounts, net settled revenue. Against that NET figure calculate channel food cost, then add packaging weighed and costed for real, not estimated. Most owners discover here that delivery food cost sits between 38% and 44%, nowhere near the 30% they believed. That number, not intuition, drives everything that follows.
Cut to 18-24 items: the ones that travel, share inventory and leave at least 68% contribution margin on net. Set app prices 15% to 25% above dine-in, benchmarked against direct competitors in your zone. Keep the PHYSICAL menu in the dining room and inside every delivery bag, and use the QR as a complement for direct ordering, price updates and analytics: print controls narrative and suggestive selling, the QR controls the data.
Build a delivery assembly station with its own mise en place, its own printer and one accountable person during high-volume windows. Track dispatch time per order daily for two weeks: past a 12-minute average the app punishes your visibility and you pay in listing position. And block the app during the window when your dining room is packed, because a table beats an order on margin every single time.
With the kitchen in order, launch a virtual brand on the same inventory — wings, bowls, sandwiches, whatever shares 70% of your inputs — and measure it as a separate business for six weeks. In parallel, drop the printed menu into every bag with a direct-order offer and a concrete incentive. Target: move app dependence from 90% to 60% within two quarters, with 1,200 owned contacts banked.
And with AI?
Optimize channels, pricing and unit economics of your dark kitchen. Diego F. Parra is an expert in AI applied to restaurants.
Free tools to apply this now
Method tools for costing the channel
None of these three replaces judgment, but all three end the argument: they turn «I feel Rappi is killing me» into a figure you can defend in front of a partner or a bank.
Use them in order: model the channel as a business first, project growth with the corrected mix second, and finish with the cash flow that tells you whether you survive the 90-day transition.
Diego F. Parra — International consultant, expert in creating and scaling restaurants and in AI applied to restaurants, foodtech and HORECA. Methodology applied in 8.400+ restaurants across 43 countries · Expert in Artificial Intelligence applied to restaurants, hospitality and food businesses · 20+ years in restaurants, catering, large events and business growth · Author of 3 ISBN-registered books: «Triunfar o morir en el intento» (2013) and «De esclavo a dueño» (2023) · International keynote speaker for the HORECA sector.
Questions that always come up about selling on Rappi
How much commission does Rappi charge a restaurant in 2026?
How much commission does Rappi charge a restaurant in 2026?
Commission runs from 18% to 30% of menu price depending on country, plan and whether you use the platform courier fleet or your own. Add co-funded campaign discounts on top and effective withholding can exceed 35% during promotional weeks.
Should I charge more in the app than in the dining room?
Should I charge more in the app than in the dining room?
Yes, and it is not gouging: it is channel costing. App pricing 15% to 25% above dine-in covers commission and packaging without touching margin. Delivery customers benchmark against other delivery, not your printed menu; real friction only appears once the gap passes 30%.
Should I open a dark kitchen from scratch or use my current kitchen?
Should I open a dark kitchen from scratch or use my current kitchen?
Start with a virtual brand on your current kitchen: near-zero added fixed cost and a clear read in six weeks. A dark kitchen from scratch only earns its keep once the virtual brand saturates installed capacity and channel contribution margin holds above 25%.
What if 90% of my restaurant depends on the apps?
What if 90% of my restaurant depends on the apps?
Drive that dependence down to 60% over two quarters. Put the printed menu in every bag with a direct-order QR, activate in-store pickup with your own incentive and work your WhatsApp base. Every point you shift from app to owned channel returns 18 to 26 margin points on the same dish.
Sector data 2026 (official sources)
Verifiable industry benchmarks from official, non-commercial sources (government, industry associations, market research) - not competitors.
| Metric | Benchmark 2026 | Source |
|---|---|---|
| Proyección de entrega de paquetes por dron a 2030 | USD 5.238,8 millones (CAGR 38,7%) | Grand View Research — Drone Package Delivery Market 2030 |
| Entregas comerciales por dron de Zipline (abril 2024) | 1 millón (primera empresa en lograrlo) | Grand View Research — Drone Package Delivery Market |
| Unidades de drones de reparto proyectadas 2024 a 2030 | de 32.456 a 275.703 unidades | Grand View Research — Drone Package Delivery Market |
| Cuota del delivery de comida en el mercado de drones 2024 | 36,87% | Grand View Research — Drone Package Delivery Market 2024 |
| Pedidos de DoorDash en el cuarto trimestre de 2024 | 685 millones (+19% interanual) | DoorDash — Q4 y Full Year 2024 Financial Results |
| Marketplace GOV de DoorDash en el cuarto trimestre de 2024 | USD 21.300 millones (+21%) | DoorDash — Q4 y Full Year 2024 Financial Results |
Related content
Grow your restaurant with the Masterestaurant method
Applied in +8.400 restaurants across 43 countries.