The chef-owned direct channel: the myth of divorcing aggregators, and what the 2026 numbers actually say

Finding 1 — At what order volume does the direct channel beat the aggregator?

The break-even shows up when the fixed monthly cost of a direct channel drops below the accumulated variable commission, and with public figures that happens sooner than almost any chef calculates.

An independent operator on Uber Eats pays a per-order commission according to eLogii (2024), and the effective cost of third-party delivery climbs once service fees and co-funded promotions are added in. Take that Bogotá chef with 1,180 orders at a 38,000-peso average ticket: 44.8 million in gross sales, of which between 13.4 and 17.9 million vanish into effective commission. The arithmetic leaves no room for debate: above roughly 400 orders a month the fixed cost wins, and wins ugly.

Finding 2 — The direct channel does not erase the cost, it moves it to another ledger line

Anyone promising you zero commission is selling smoke in elegant handwriting. Typical third-party platform commission does not evaporate when you launch your own site: it migrates into traffic acquisition, support, packaging, couriers and refunds. The real difference lies in the NATURE of the spend. Commission is a toll charged on every transaction and it rises in December exactly as your sales do; paid media and local SEO are investment you switch off, raise or reassign on a Tuesday morning. When a chef tells me he wants to save the commission, I correct the verb: he does not save it, he CONTROLS it. And controlling a 12% cost is a different business from suffering a 35% one.

Finding 3 — Platform concentration is territory risk, not consultant rhetoric

Two companies take more than 95% of India's online delivery —Zomato and Swiggy, per Business of Apps (2025)— while in China the combined share of Meituan and Ele.me tops 90% (Mordor Intelligence, 2025). Meituan alone reported over 14.5 million active merchants in 2024. That is the endgame of any mature delivery market: a duopoly that sets the price of being visible. Colombia is already well down that road, with Rappi declaring more than 30,000 partner merchants and 7 million monthly orders (La República / Rappi, 2024). Ask yourself what happens if tomorrow your category takes a three-point commission hike and you hold neither a customer base nor a direct channel: you do not negotiate, you comply. The chef who never built his own list discovers his brand was never his, just a rented line in somebody else's search results.

Finding 4 — What each peso buys you: rented attention or accumulated asset

The aggregator sells attention in cash and the direct channel builds equity, and that is the difference no monthly spreadsheet ever captures. An order that cost 35% in effective commission (ActiveMenus) disappears the moment it is delivered: no email, no phone, no contact permission, no review that belongs to you. A Google Business Profile listing with photographs of the actual plate and five-star reviews does the opposite: every month it weighs a little more on the local map and it keeps producing after you decide to leave the platform. With over a million partner merchants on Uber Eats and close to 95 million users (Uber Technologies, 2024), your competitive edge will never be presence there, because everybody is there. It will be someone searching for you by name rather than by category.

Finding 5 — How the numbers behave once volume climbs

Here sits the trap that costs the most: a business with its own channel improves margin as it grows, and a business dependent on aggregators worsens it. Add labor cost, which the U.S. Bureau of Labor Statistics places between 25% and 35% of revenue, and you will see why so many virtual brands bill more each quarter and earn less. Growth on borrowed channels is a machine for manufacturing volume without profit. Doubling sales to earn the same is not scaling; it is running faster inside the same cage.

Finding 6 — Your own pickup is the fastest lever and almost nobody pulls it

If you need a result within thirty days, start with in-store pickup rather than delivery. Those twenty-odd points of difference are gross margin requiring not one new courier nor an app of your own. The operational play is simple and most people dismiss it for being unglamorous: turn the pickup window into a fast, well-signposted experience with hours you actually keep, and push there the customers who already live eight blocks away.

Finding 7 — Which sources hold this analysis up and what the Masterestaurant reading adds

The numbers in this study belong to whoever published them and the interpretation is mine. Bureau of Labor Statistics form the base. On top of it, Diego F. Parra and the Masterestaurant method apply a single unit-economics question: which decision triggers each figure, and at what volume does it flip sign. I got this wrong for years, recommending operators walk away from aggregators; today I argue the opposite. Stay, collect the volume they bring, and use every order to capture a customer who next time arrives direct.

Finding 8 — The operating decision: coexistence with rules, not divorce

The direct channel does not replace the aggregator, it UNLOADS it, and that distinction governs the entire execution. Set a ceiling: no operator should hold more than 50% of sales on a single platform charging between 15% and 30% nominal (Food On Demand, 2026), because that threshold is where negotiation ends and dependency begins. With 86% of operators declaring themselves comfortable using artificial intelligence (Toast, 2025), automating customer data capture and post-sale follow-up is no longer a twelve-month project nor does it demand a technical team. This week, do one measurable thing: print a QR code with an exclusive offer on every package leaving through the aggregator and count how many customers return through your channel within thirty days. That number, not the theory, will tell you what your brand is worth off the platform.

Finding 9 — Three differences that decide the mix

FIRST: aggregators charge variable, direct channels charge fixed. So the right question was never which one is cheaper, but at how many orders the fixed cost beats the variable one. SECOND: the aggregator rents attention while the direct channel builds an asset. Third-party delivery commission leaves nothing behind; a Google Business Profile with 5★ reviews accumulates position month after month and stays yours even if you switch platforms. THIRD: concentration is genuine territory risk, not a consultant's ghost story. In India, Zomato and Swiggy exceed 95% of online delivery (Business of Apps, 2025); in China, Meituan and Ele.me pass 90% of orders (Mordor Intelligence, 2025). When two players own the channel, next year's commission is their decision. A chef without a direct channel does not negotiate. He complies.