The Silent Commission: What Each Aggregator Order Really Costs You

1. How much does each aggregator order really cost you?

The contract commission (25%–30%) is the starting point, not the real cost: once you add food cost, premium packaging, in-app advertising, and the operational surcharge of the rush, the order can consume more than half the ticket.

Delivery is now revenue infrastructure, not a marginal channel: the global delivery app market hit USD 110 billion in 2024 (+15.5%) per Business of Apps (2025). The problem isn't selling; it's whether margin survives. I see it in dozens of digital menus: the owner watches the dollars coming in from Rappi or iFood, not the contribution margin per order. If your dish carries a 30% food cost, commission takes another 28%, and packaging 5%, you're already at 63% before touching in-app advertising. The executive question for 2026 isn't how much you bill on the aggregator: it's how much you keep per order after subtracting everything.

2. Commission is only the first deduction, not the last

The visible 25%–30% commission is merely the first cut in a cascade that is rarely modeled in full. In practice, each order stacks: aggregator commission, dish food cost, transport-ready packaging (premium in delivery, not optional), in-app advertising to avoid sinking to the bottom of the listing, and the operational surcharge of the peak —overtime, waste, and errors when the kitchen saturates—. That stacking is exactly what fuels the platforms' business: DoorDash's marketplace GOV closed Q4 2024 at USD 21.3 billion, +21% per DoorDash (Q4 and Full Year 2024). Their growth is your cost variability. The mistake I see over and over is treating commission as the total cost; the total cost is the sum of the cascade. A dish that leaves a 68% gross margin in the dining room can drop to 12% through the same digital channel if nobody ran the math order by order.

3. The real KPI isn't commission: it's contribution margin per order

The KPI that decides whether delivery adds or drains EBITDA is contribution margin per order, not the commission percentage. Contribution margin = dish price − (food cost + commission + packaging + prorated advertising fee). That number, not gross sales, tells you which orders are worth pushing. The market won't shrink: China projects USD 539.87 billion in food delivery revenue in 2026 per Statista (2026), and Mexico's online delivery alone reached USD 9.22 billion in 2024 (CAGR 14.66%) per Statista (2024). With that demand, the risk isn't lacking volume: it's having volume without margin. In the Masterestaurant method we measure each digital SKU by its contribution in dollars, not in percentage: a dish with an 8% margin but USD 4 of contribution can beat one at 20% that leaves USD 1.50. Push the dollars that remain, not the percentage that looks pretty. A profitable digital menu is designed with delivery menu engineering; you don't list the dining-room menu with food cost out of control.

4. Design a delivery-specific menu, don't clone the dining room

The dish that travels 25 minutes, is served in packaging, and competes on price in an app is not the same one that goes straight to the table. Diego F. Parra puts it plainly: in delivery, the expensive ingredient that doesn't survive transport is money you pay twice. Dark kitchen logic already moves serious capital: the cloud kitchen market projects USD 83.5 billion in 2026, CAGR 9.7% to 2034 per Fortune Business Insights (2026), and Mexico alone added USD 1.1 billion in 2024, CAGR 10.74% toward 2033 per IMARC Group (2024). The cash rule is hard: dish food cost ≤ 32% as a maximum, never as a target. Redesign portions, packaging, and price so contribution survives the commission; if a dish can't withstand the digital channel, it doesn't belong on the digital menu. Every virtual brand must be measured by its own break-even point, not by the operation's average.

5. Every virtual brand has its own break-even point

Launching three ghost brands from one kitchen sounds like leveraging assets, but if you don't separate costs per brand, some add EBITDA while others bleed silently under the consolidated figure. Market appetite is real and regional: Asia-Pacific ghost kitchens were worth USD 21.73 billion in 2024, heading to USD 60.59 billion in 2032 (CAGR 12.8%) per Coherent Market Insights (2024); Spain recorded USD 928.22 million in ghost kitchens in 2023 per Expert Market Research (2024). In the Masterestaurant method each virtual brand carries its own food cost, commission, in-app advertising, and kitchen proration, and is required to cover its break-even. The one that doesn't reach its break-even point is corrected or closed; it isn't hidden behind the one that works. Consolidating without breaking down is how a channel that 'grows' makes you poorer. Reducing territory risk demands diversifying aggregators and building direct ordering, because depending on a single app means handing your margin and your data to a third party.

6. Diversify aggregators and build your own channel

The direct channel has better unit economics by design: without the 25%–30% commission, the same sale leaves far more contribution margin. The size of the pie justifies fighting for it: Brazil moved close to USD 18.8 billion in online delivery in 2024, the largest in Latin America per Statista (2024), and Colombia USD 1.18 billion in 2024, CAGR 7.32% (2024-2029) per Statista Market Insights (2024). The concrete action: use aggregators to discover demand and capture new customers, but migrate the recurring ones to your direct ordering with incentives that cost less than the commission you save. A customer who orders through your own channel twice a month already pays their acquisition cost; the rest is margin the platform used to keep. Diversify channels the way you diversify suppliers: so you're hostage to none. You stop treating commission as the cost and start treating contribution margin per order as the real KPI.

7. What changes when you govern delivery unit economics

You design a delivery-specific digital menu (menu engineering) instead of listing the same dining-room dishes with out-of-control food cost. You measure each virtual brand against its own break-even: you discover which add EBITDA and which bleed you silently. You reduce territory risk by diversifying aggregators and building a direct channel (direct orders with better unit economics).