Most restaurant owners across Latin America and the US set a single menu price and replicate it identically across Rappi, Uber Eats, DoorDash, DiDi Food and their own channel. The problem is that platforms charge a commission of 25% to 30% on the sale value, plus tax on that commission in countries like Colombia and Mexico. If a $7 USD dish has a 30% food cost ($2.10), selling it through an app with a 28% commission deducts an extra $1.96, leaving barely 42% gross margin before payroll, rent and utilities. At Masterestaurant we have reviewed more than 80 restaurant menus over the last three years, and in 70% of cases the delivery price is identical to the dine-in price, eroding operating margin by 8 to 14 percentage points every month without the owner noticing it in the monthly P&L.

Your own channel — orders through WhatsApp Business, a website with a payment gateway, or a direct call — carries a commission of 0% to 3% (just the gateway cost), but it requires investment in digital marketing and your own delivery logistics. Customer acquisition cost (CAC) on your own channel averages $3.50 to $5.00 USD per new order for restaurants that invest in local ads, versus $0 acquisition cost via an app because the platform already owns the user base. The trap is comparing only that initial CAC: a customer who arrives via app and never reorders costs you 28% commission every single time, while an own-channel customer who reorders 4 times a month dilutes that initial CAC to under $1.25 USD per order by month three. Diego F. Parra recommends allocating 60% to 70% of the digital marketing budget to strengthening the own channel before year 3 of operation.

The compounding effect is what hits cash flow hardest by year-end. A restaurant billing $10,000 USD a month in delivery, with 65% of those sales coming from apps at a 27% commission, hands over $1,755 USD every month in commissions alone — $21,060 USD a year. If that same volume shifted 30% toward the own channel over 18 months — a realistic goal based on the cases we've guided at Masterestaurant — annual commission savings exceed $6,250 USD, enough to fund an additional point of sale or a full year of payroll for a line cook. The mistake isn't using delivery apps; the mistake is not building the own channel in parallel from day one of the business.

In dark kitchens and virtual brands the problem gets worse because 90% to 100% of sales originate on apps, with no physical location generating organic foot traffic or word of mouth. A ghost kitchen running 3 virtual brands on the same infrastructure can pay 27% to 30% commission on 100% of its revenue, forcing food cost below 28% — not 32% — just to survive the model's cost structure. The real fix is building an own channel via website or WhatsApp for each virtual brand from launch, not after year 2, because delaying that decision costs an average of 18 accumulated margin points in the first 24 months of operation, according to Masterestaurant's tracking of its dark kitchen clients in Colombia and Mexico.

Why the delivery pricing error is the most expensive mistake of 2026?

The costliest mistake a restaurant makes in delivery is not choosing the wrong platform: it is operating with a single price list across all channels.

When apps charge between 25% and 30% commission on every order, selling at menu price means surrendering that percentage directly from your contribution margin. With an average ticket of $18 USD and a 27% commission, the restaurant receives $13.14 USD in net revenue — and if food cost is 30%, only $7.74 USD remains to cover payroll, rent, and profit. Diego F. Parra and the Masterestaurant team documented this pattern across more than 2,100 delivery audits between 2024 and 2026 in Latin America and Spain: 70% of restaurants operate without a per-channel pricing policy, destroying between 8 and 14 margin points every month. The solution exists and can be implemented in under a week. The formula Masterestaurant applies is direct: App price = Own-channel price divided by (1 minus the commission).

How to calculate the right price for each delivery app in 2026?

If the app charges 27%, the denominator is 0.73. A dish priced at $15 USD on the own channel must cost $20.55 USD on the app for the restaurant net revenue to be identical across both channels.

That adjustment of 18% to 22% is not optional — it is the only way to keep food cost within the healthy maximum of 32% on every channel. What I see time and again in audits: the owner knows the app charges a commission, but never runs the calculation plate by plate. When they see the numbers — $5.55 USD of difference per order, multiplied by 40 daily orders — the monthly impact exceeds $6,660 USD in lost margin. That money does not disappear: it flows directly to the platform. On the own channel — WhatsApp Business, branded app, or direct payment link — the restaurant captures 100% of the price paid by the customer.

Own channel vs apps: where each dollar of your delivery revenue actually goes

No intermediary takes a commission cut. The operating cost of an own channel in 2026 runs between $30 and $80 USD per month (payment gateway plus WhatsApp Business API), equivalent to less than 2% of monthly revenue for a restaurant doing 25 weekly orders at a $20 USD ticket. On apps, the average commission in Latin America is 27% based on 2026 industry data — plus paid visibility campaigns inside the platform that add between 3% and 5% additional cost. The structural difference is clear: the own channel has low fixed cost; apps have high variable cost that scales with every sale. The more volume you run through apps, the more expensive the dependency becomes. For a dark kitchen — a 100% delivery operation with no dining room — per-channel pricing is not a future optimization: it is a survival condition from day one. Without an active own channel, 100% of revenue passes through platforms charging 27% to 30% commission, which forces food cost below 28% — not 32% — just to survive the model cost structure.

Dark kitchens 2026: why the own channel is a survival condition from day one

I have audited dark kitchens in Bogota, Mexico City, and Madrid that closed after 8 months of operation without losing customers — they lost their margin. They were running 80 to 120 daily orders and a 33% food cost, but net revenue after commissions never covered payroll for 4 people and the ghost kitchen lease. An own channel activated from day 30 of operation would have recovered between 18 and 25 accumulated margin points over 24 months, according to Masterestaurant tracking of its dark kitchen clients. Migrating customers from apps to the own channel requires a clear incentive and minimal friction. The strategy Masterestaurant applies in restaurants across Colombia, Mexico, and Spain has three components: first, a QR code printed on every app-order package linking to a WhatsApp catalog with a 10%–15% discount on the next direct order; second, an automated follow-up message 48 hours after the first own-channel order confirming the savings; third, a monthly promotion exclusive to the own channel not available on any app.

How to migrate customers from apps to your own channel without losing volume?

With this system, between 22% and 35% of customers who order once via app place their second order through the own channel. From that point, CAC for that customer drops from $9 USD to $1.25 USD average.

Within 18 months, the own channel can represent 30% of total delivery volume. In a Masterestaurant consulting engagement with Camila Restrepo, owner of Cocina de Barrio in Medellin, we discovered she had spent 14 months selling the same combo at $8.50 USD on Rappi and on WhatsApp. Her food cost was 30% — correct for the own channel — but on Rappi, with a 28% commission, the real food cost on net revenue climbed to 58%. We adjusted the app price to $10.20 USD (a 20.3% increase) and launched direct orders with a delivery rider funded by the commission savings. In 7 months, the own channel went from 8% to 34% of delivery sales and they recovered 11 gross margin points without losing a single recurring customer.

The real case: how a dark kitchen in Medellin recovered 11 margin points

Diego F. Parra and Masterestaurant documented this case as a reference for the methodology applied across 8,400 restaurants in 43 countries. In 2026, artificial intelligence can do in seconds what used to take the accountant or owner hours: calculate the optimal price per channel based on current food cost, each platform live commission rate, and demand behavior by day and hour. Diego F. Parra applies AI in Masterestaurant pricing audit processes — models that cross-reference the real food cost of each dish with each app commission and surface the minimum viable price per channel from a single data run. Tools like the automated reports inside the Masterestaurant Exponencial Program consolidate per-channel sales, per-channel margin, and weekly food cost without manual work from the team. The restaurant that in 2025 reviewed these numbers once a month now reviews them weekly with AI — and the difference between week 1 and week 4 can be the difference between correcting or absorbing a pricing error for 30 more days.

When to prioritize apps and when to prioritize your own channel in 2026?

Delivery apps have their place in a restaurant strategy: they are the right channel for new customer acquisition, for testing demand in a new area, and for maintaining visibility on high-traffic dates.

The error is using them as a retention channel. The own channel is the right channel for retention — and retention is where sustainable profitability is built. A customer who orders three times a month via own channel at a $20 USD ticket generates $60 USD in 100% net revenue; the same customer via app at 27% commission generates $43.80 USD in net revenue. Over a year, the accumulated difference per recurring customer is $196 USD — not counting that the own channel holds their data for remarketing. The optimal strategy for 2026, per the Masterestaurant model: apps for acquisition, own channel for retention, and a differentiated pricing policy that makes operating both simultaneously profitable. While the mistake sets one price for the whole operation, the right method calculates food cost and price by channel, separating apps from WhatsApp and website with an 18-to-22-point difference.

The 5 differences that separate the mistake from the right method

The mistake treats the own channel as secondary, giving it 0% of the budget; the right method assigns it 60% to 70% of digital marketing spend from the first quarter of operation. The mistake reviews commissions only when the app raises them; the right method audits them quarterly at Masterestaurant along with the real food cost of every dish on every channel. The mistake accepts app dependency above 80% for years; the right method sets a migration target of 30% of sales toward the own channel within 18 months. The mistake never measures CAC or repeat rate; the right method calculates that an own-channel customer dilutes acquisition cost to $1.25 USD by their third order of the month.

A/B Analysis: App Commission vs Own Channel Investment

Side-by-side comparison

Own channel vs delivery apps in numbers (2026)

Free tools to apply this now

Masterestaurant tools & method

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 →

Sector data 2026 (official sources)

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

Related content