Delivery Commissions That Kill Your Margin: Before vs After with Masterestaurant

In 2026, delivery represents an average of 38% of total sales for urban restaurants in Latin America, according to foodtech industry data. But that 38% carries an average commission of 29.5% on the order value, before VAT, before packaging, before any discount the platform applies unilaterally during promotional campaigns. The mathematical result is brutal: a restaurant with a 15 USD average ticket and 30% food cost loses money on every order if the commission exceeds 27%. And in most markets, it never goes below that.

The mistake I see time and again is running the delivery channel with the same menu and prices as the dining room. That menu was designed for a cost structure where payroll and rent are already covered by in-house traffic. When that same dish travels through Rappi or Uber Eats, it carries a 30% commission, packaging cost of 0.80 to 1.50 USD per order, and image degradation because it arrives cold. Diego F. Parra, Masterestaurant.

Why a 30% commission destroys margin on every single order

A 30% commission on order value is not a marketing cost: it is a silent partner taking 3 out of every 10 dollars your restaurant bills through delivery. In 2026, the average commission from leading platforms in Latin America sits at 29.5%, but the real cost is higher because platforms apply unilateral discounts during campaigns, between 10% and 20% additional, deducted from the amount transferred to you, not from what the customer pays. For a restaurant with a 15 USD average ticket and 30% food cost, that means the margin before any operating expense is negative. Diego F. Parra, Masterestaurant, has analyzed more than 60 operations across LATAM where the delivery channel represents between 35% and 55% of sales and the net margin on that channel is below 3%. The problem is not delivery: it is operating that channel without having designed for it. Your dining room menu was designed for a cost structure where payroll and rent are already distributed across in-house diners.

The mistake of using your dining room menu on delivery apps

When that same menu lands on Rappi or Uber Eats, it carries a 30% commission, packaging cost between 0.80 and 1.50 USD per order, and the experience degradation of a crispy item arriving soggy and a salad arriving oxidized. The actual food cost on delivery without reformulation ranges from 32% to 36%, based on operations audited by Masterestaurant in 2025-2026. Reformulating for delivery means eliminating items that travel poorly, fragile textures and proteins that lose moisture in 20 minutes, and redesigning star recipes with higher-yield starches and incorporated sauces that stabilize the dish. The measured result is a 6 to 8 point food cost reduction without reducing the perceived portion size for the customer. Delivery platforms in 2026 have commission tier structures ranging from 18% to 32%, but 73% of Latin American operators have never initiated a negotiation. The mistaken assumption is that there is no bargaining power.

How to negotiate a lower commission tier with data in hand

The reality is that a restaurant with more than 300 monthly orders and a cancellation rate below 3% has solid arguments for requesting the lower tier. The conversation with the account manager must include three data points: current volume in units, average ticket, and a 20% growth projection over 60 days conditioned on the new tier. That projection is not arbitrary, it comes from the menu redesign and app price adjustment already implemented. Rappi and Uber Eats in Colombia, Mexico, and Brazil can lower commissions by 5 to 8 points under those conditions. Diego F. Parra, Masterestaurant, has documented this in 12 successful negotiations between 2025 and 2026. Raising prices on delivery apps is the fastest lever and the one that generates the most psychological resistance in operators. The data point that changes the conversation: delivery customers accept paying between 12% and 18% more than the dining room price if the delivery experience is consistent, with intact packaging, correct temperature, and delivery time under 35 minutes.

Adjusting prices on the app: how much to raise without losing volume

78% of regular delivery customers do not switch restaurants after a 15% price increase, according to behavioral analysis on platforms in Mexico and Colombia during 2025. The mechanics are direct: if the average ticket rises from 14.20 USD to 16.30 USD with a 15% adjustment, and commission is 30%, the net income per order goes from 9.94 USD to 11.41 USD, a 14.8% increase without changing a single recipe. That differential, accumulated over 200 monthly orders, equals 294 USD of additional margin per month per location. The proprietary channel, including WhatsApp Business, order landing pages, and proprietary apps, does not compete with the platforms: it complements them. The correct logic in 2026 is to use Rappi or Uber Eats as an acquisition channel for the customer who does not know you yet, and migrate the repeat customer to a channel where commission is zero. The most effective and lowest-cost mechanism is the physical insert in the order: a 0.05 USD card with a QR to WhatsApp Business and an offer of 2 USD off the third direct order.

Proprietary channel: how to move 20% of volume without losing customers

The conversion rate for customers with two prior orders is 28%, based on data from 8 operations implemented by Masterestaurant in 2025. Within 90 days, a restaurant doing 200 orders per month can have 50-60 active customers on the proprietary channel. Those 50 customers generate 750-900 USD of monthly revenue with zero commission, equivalent to recovering 4.5 margin points across the total delivery channel. Packaging is the expense no delivery operator controls because it seems small at the unit level. In a restaurant doing 200 monthly orders, the difference between 1.40 USD generic packaging and a 0.75 USD standardized kit equals 130 USD per month: 1,560 USD per year per location. For a 3-location chain, that is nearly 4,700 USD of recoverable margin without changing a recipe or negotiating with anyone. The Masterestaurant standard delivery kit includes three pieces: a main container with hermetic seal at 0.35 USD, a first-use thermal bag at 0.25 USD, and a proprietary channel card insert at 0.05 USD.

The invisible packaging cost: 1,560 USD per year per location nobody sees

Total: 0.65 USD at a volume of 600 monthly units. The presentation difference versus 1.40 USD generic packaging is minimal in customer perception but massive on the P&L at year-end. The weekly report from Rappi or Uber Eats shows gross sales, applied discounts, and the net transfer. What it does not show, and what actually matters, is the real margin after food cost and packaging. The correct analysis takes 45 minutes and uses three data points: net platform transfer for the period, total food cost for orders delivered calculated as recipe grams multiplied by ingredient price, and total packaging cost. The difference between net transfer and the sum of food cost plus packaging is the gross delivery margin before any operating expense. If that number is below 35% of gross sales, the channel is in the risk zone. If it is below 20%, there is real loss even if the income statement does not show it, because the dining room fixed costs are subsidizing it.

How to read the platform report to know if the channel is destroying margin

Diego F. Parra, Masterestaurant, calls this the volume illusion: selling more on delivery while the dining room covers the channel losses. The most frequent scenario Masterestaurant finds in restaurants with over 40% of sales through delivery is this: the P&L shows a net profit of 6-8%, but that result only exists because the dining room generates an 18-22% margin that compensates the negative margin of the digital channel. Diego F. Parra calls this cross-channel subsidy. The danger is not immediate, the restaurant functions, but the structural risk is severe: if the dining room drops 20% due to any external event, the delivery channel has nobody left to subsidize it and the operation enters negative cash flow in under 6 weeks. The solution is not to reduce delivery volume: it is to make that channel self-sufficient, with food cost at or below 26%, negotiated commission at or below 25%, and at least 15% of volume on a proprietary channel.

What happens when the dining room subsidizes delivery losses

Those three numbers together generate a self-sustaining delivery margin of 9% to 14%, without depending on the dining room. The delivery menu is not the same as the dining room menu. A dish designed to travel in a polystyrene container has to survive 25 minutes at 136 degrees Fahrenheit and still be appealing. That forces recipe reformulation: fewer crispy textures that lose temperature, more incorporated sauces that travel well. And that reformulation, done with criteria, drops the food cost between 4 and 7 points by eliminating fragile garnishes and reducing protein portions offset by higher-yield starch. Diego F. Parra, Masterestaurant, documents this in more than 60 restaurants intervened since 2022. The 30% commission is not fixed: it is the price of not negotiating. Platforms in 2026 offer tiers ranging from 18% to 32% based on monthly order volume and account seniority. A restaurant billing more than 8,000 USD per month on the app has real leverage to request the next tier down.

The 4 differences that move the margin in delivery

The problem is that 73% of operators never initiate that conversation because they believe they cannot. The proprietary channel does not compete with the apps, it complements them. The strategy that works in 2026 is using the app for discovery, where the customer finds you for the first time, and migrating the repeat customer to WhatsApp Business with a 2 USD discount on their third direct order. That customer on the proprietary channel has an acquisition cost of 0 USD versus the 4.50 USD implicit in the platform commission. Packaging is the invisible cost nobody controls. In a restaurant doing 200 orders per month, the difference between 1.40 USD and 0.75 USD packaging equals 130 USD per month, or 1,560 USD per year. Multiplied across 3 locations, that is nearly 4,700 USD of recoverable margin without changing a single recipe.

Before vs after analysis: the 5 margin levers in delivery

Side-by-side comparison

2026 data: the real cost of delivery without strategy

4 steps to stop losing margin on delivery

Free tools to apply this now

MASTERESTAURANT tools to control delivery

The MASTERESTAURANT method for delivery has three pillars: channel diagnosis, menu redesign, and proprietary channel activation. None of the three requires expensive technology investment or changing platforms. They require data, judgment, and execution in that order.

Diego F. Parra, Masterestaurant, works with three proprietary tools that allow this process to be done in weeks, not months, with real cash data rather than industry averages that do not apply to your specific operation.

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

Sector data 2026 (official sources)

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

Related content