Delivery commissions that kill the margin: myth vs reality

The myth was born on social media and restaurant-owner WhatsApp groups, where the figure '50% of the order' keeps circulating. The real number in the US, Mexico and Colombia in 2026 is different: standard commissions from Uber Eats, DoorDash and Rappi run between 27% and 30% of the ticket, according to current marketplace contracts. On top of that, some restaurants add 3%-5% more when using the platform's own courier for short-distance orders, or 1%-3% for in-app advertising (boosts). The miscalculation comes from stacking commission, tax on the commission, and payment-processing cost without separating line items, which inflates the perceived rate to 40%-45%. Diego F. Parra, from Masterestaurant, insists that before negotiating with any platform, owners should request a line-by-line breakdown of the last billing cycle: base commission, marketing, logistics and withholdings. Without that breakdown, any conversation about margin is a conversation in the dark.

The math is harsher than the myth, and easier to fix. Take a dish with a 30% food cost, within the 32% ceiling recommended by the Masterestaurant method. Add the platform commission (28% average in 2026), disposable packaging (4%), and the digital payment fee (2%). Direct order costs reach 64% of the ticket, before payroll, rent or utilities, which the break-even model charges separately. That leaves 36 points to cover fixed costs and profit. If the restaurant charges the same price as in the dining room, that 36% usually isn't enough, and the delivery dish runs an operating loss of 2% to 5%. The real fix isn't cutting food cost below a safe level; it's raising the platform price by 10%-15%, which recovers 4 to 6 margin points without touching dish quality or target food cost.

Dark kitchens change the equation because they remove dining-room cost. A traditional location spends 8%-12% of the ticket on rent and another 25%-30% on front-of-house payroll that a 100% delivery model doesn't need. That structure absorbs a 27%-30% commission without collapsing the business, because production-kitchen rent runs 40%-60% lower than a full dining-room location in cities like Miami, Mexico City or Bogotá. Diego F. Parra has documented dark kitchens running a 31% food cost and a 29% commission that still close the month with 5%-7% net margin, thanks to that structural savings. The risk sits with the hybrid restaurant, the one keeping a full dining room and selling delivery at the same price: there's no rent savings to offset the commission, and the digital channel ends up subsidized by the physical operation without the owner noticing it on the P&L.

The error repeated across most Masterestaurant audits isn't the commission, it's the absence of a delivery-specific menu. A menu priced 25%-30% higher than dine-in, justified by packaging, commission and logistics, is standard in mature markets like the US and Spain, but in Latin America only 3 out of 10 restaurants apply it, according to last year's cash reviews. The other 7 compete on in-app price, run 20%-30% discounts without measuring the resulting food cost, and end up selling dishes with a real food cost of 38%-42% once packaging and commission are added. That's the true margin killer: not the platform's fee, but the lack of a price and recipe built specifically for the digital channel. Fixing it doesn't require leaving the apps; it requires a parallel menu with its own cost engineering.

How much do Rappi, Uber Eats, or DiDi Food actually charge in 2026?

The standard commission charged by the three leading platforms in Colombia, Mexico, and Chile ranges between 27% and 30% of the ticket value in 2026 — not the 45%-50% figure circulating in restaurant-owner WhatsApp groups. That myth was born from stacking the base commission, VAT on the commission, and the digital payment fee without separating each line item, which inflates the perceived total to 40%-45%. On top of the base commission, platforms may add 3%-5% if the restaurant uses the platform's own delivery rider, and 1%-3% if in-app advertising boosts are activated. Diego F. Parra of Masterestaurant recommends requesting a line-by-line billing breakdown for the most recent month before any negotiation: base commission, logistics, marketing, and withholdings. Without that breakdown, debating whether the platform 'kills the margin' is a guessing game; with it, most restaurants discover that 60%-70% of total platform cost comes from just the first two line items — which is where real negotiation leverage lives.

What actually kills the margin in delivery: the commission or the pricing?

The real margin killer is not the platform fee — it is charging the same in-app price as in the dining room while running a food cost that ignores packaging and commission. Take a dish with a 30% food cost, within the 32% ceiling set by the Masterestaurant method. Add the platform commission (28% average in 2026), disposable packaging (4%), and the digital payment fee (2%): direct order costs reach 64% of the ticket before touching rent, payroll, or utilities. That leaves 36 points to cover fixed overhead and generate profit. When the app price matches the dine-in price, that 36% rarely covers fixed costs, and the delivery dish closes with an operating loss of 2%-5%. The fix does not require cutting quality or dropping the food cost below safe levels; it requires raising the app price by 10%-15%, recovering 4 to 6 margin points without changing the recipe.

How much higher should app prices be compared to dine-in prices?

App prices should be 10%-15% higher than dine-in prices as a baseline rule under the Masterestaurant 2026 method. That differential is not arbitrary: it covers the incremental commission versus a waitstaff cost (platforms charge 28%; a dining room allocates 8%-12% to service), the disposable packaging expense (3%-5%), and a safety buffer for in-app promotions so the dish never crosses into negative margin. In mature markets like the United States and Spain this adjustment has been standard practice for years, but in Latin America only 3 out of 10 restaurants apply it according to cash-register audits from the past year. The other 7 compete on price inside the app, activate 20%-30% discounts without measuring the resulting food cost, and end up selling with a real food cost — once packaging and commission are added — of 38%-42%, a range where no restaurant sustains operations at mid-term without eroding its capital base.

Is it worth negotiating the commission with the platform?

Yes, commissions are negotiable, and volume is the only argument that moves the needle. With more than 200 consistent monthly orders, account managers at Rappi and Uber Eats open negotiations; the typical reduction is 4 to 6 percentage points off the base commission, which on a 25 USD average ticket represents savings of 1 to 1.50 USD per order — or between 200 and 300 USD per month at that minimum volume. What is not negotiable in most 2026 marketplace logistics contracts is the digital payment fee (1.5%-3%) and the VAT on the commission, which in Colombia applies at 19% and in Mexico at 16%. The most common error Masterestaurant documents in digital channel audits is that the owner negotiates the commission percentage without first calculating the break-even point per order: they drop 3 commission points but raise food cost 20% through recipe changes, and the net result is worse than before the negotiation.

What real advantage does a dark kitchen have over a traditional restaurant?

A dark kitchen eliminates the costs a dining-room location cannot compress: rent (8%-12% of the ticket in prime zones of Bogotá, Mexico City, or Santiago) and front-of-house payroll (25%-30% of the ticket in a 40-table model). By removing those two line items, a production kitchen can absorb the 27%-30% platform commission without collapsing. Diego F. Parra has documented dark kitchens operating with a 31% food cost and a 29% platform commission that still close the month with 5%-7% net margin, because production-space rent falls 40%-60% compared to a dine-in location in the same cities. The risk lies in the hybrid restaurant: one that maintains a full dining room while also selling via delivery at the same price, with no structural savings on rent or payroll. There, the digital channel ends up subsidized by the physical operation, and the owner only discovers it when cash flow stops balancing.

How do I calculate the real food cost of a dish for delivery?

The real food cost of a delivery dish is not the same as the kitchen food cost. Start with the ingredient cost as a percentage of the selling price; if that ratio is 28%, it sits well within the 32% ceiling Masterestaurant recommends. But for the delivery channel, add the disposable packaging cost: bags, boxes, cutlery, sauce sachets. On average, packaging represents 3%-5% of the ticket for dishes priced between 10 and 30 USD, and rises to 6%-8% for restaurants using premium branded packaging. With those two components, the real delivery food cost of a dish that costs 28% in the kitchen climbs to 31%-36%. If a 15% in-app discount is also activated — the most common promotion on Latin American platforms — the effective food cost on the net price received reaches 36%-42%, exceeding the viability threshold. Cost engineering for a delivery menu starts with the recipe and the packaging, not with the platform negotiation.

What is the first concrete step to fix the delivery channel without leaving the apps?

The first step is to build a parallel delivery menu with its own prices, packaging costs built in, and no dish whose real food cost plus commission exceeds 32% of the net price received. There is no need to leave the platforms or reduce the number of dishes; what is needed is to separate the dine-in cost engineering from the digital-channel cost engineering. The Masterestaurant method proposes doing this in four sessions: (1) download the platform's line-by-line billing breakdown for the most recent month; (2) calculate the real food cost for each delivery dish by adding ingredients and packaging; (3) adjust in-app prices so gross margin stays above 36% of the net price; (4) review results at 30 days comparing average ticket, order count, and gross margin by channel. With that cycle, most restaurants Masterestaurant has accompanied recover between 4 and 8 margin points in the first 60 days without significant loss in order volume.

5 differences between the myth and the reality of delivery commissions

Difference 1: the real commission is 27%-30%, not the 45%-50% circulating online. Difference 2: the food cost that kills margin isn't the kitchen one (32% max); it's the real food cost with packaging, which adds 3%-5% more. Difference 3: the delivery menu price should run 10%-15% higher than dine-in, not equal to it. Difference 4: the commission is negotiable; with +200 monthly orders it drops 4 to 6 points. Difference 5: a dark kitchen saves 40%-60% on rent, offsetting a commission that a dine-in location can't absorb the same way.

A/B Analysis: Negotiate commission or adjust price first?

Side-by-side comparison

Delivery commissions, by the numbers (2026)

How to protect delivery margin in 4 steps (Masterestaurant method)

Free tools to apply this now

Masterestaurant tools to control delivery margin

These three tools from the Masterestaurant ecosystem turn channel-margin math into a minutes-long process, not an improvised spreadsheet.

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 the book «From Slave to Owner» (Amazon) · International keynote speaker for the HORECA sector.

Full profile →

Frequently asked questions about delivery commissions and margin

Sector data 2026 (official sources)

Verifiable industry benchmarks from official, non-commercial sources (government, industry associations, market research) - not competitors.

Related content