Zomato and Swiggy commissions have crossed the point where the aggregator-only channel mix kills margin. Indian restaurants are quietly rebuilding the order flow. Here's the playbook.
Previous article
41% of consumers expect personalized restaurant experiences. 97% of operators are responding. The structural problem is that paper menus cannot personalize. Here's the playbook.
Next article
For most of the last decade, the Indian restaurant operator's relationship with food delivery aggregators has been a stable trade. Zomato and Swiggy bring the customers. The restaurant pays a commission. Everyone moves on. The math worked because aggregator commissions were a manageable line item against the reach they delivered.
That math has quietly broken. Aggregator commissions in India now routinely sit in the 18 to 30% range per order, climb higher during promoted-listing campaigns, and have been moving up year over year for the last four years. Layer that on top of the cost picture every Indian restaurant is already absorbing, food inflation, beef and oil price spikes, rising rent, rising labor, and the aggregator commission has crossed from "cost of doing business" into "structurally incompatible with margin."
The smart operators have noticed. They are not abandoning Zomato or Swiggy. The aggregators still drive real volume, especially for new customer acquisition. What is happening instead is more interesting and more strategic: restaurants are building a parallel direct ordering channel, training their existing customers to use it, and handing the delivery to Porter or a similar third-party logistics partner. The aggregator becomes one channel among several. The direct channel keeps the margin.
This is the most consequential operational shift Indian restaurants have made in five years, and almost nobody is writing about it.
Aggregator commissions in India work on a sliding scale based on the restaurant's negotiating leverage, the city, the cuisine category, and the promotional terms. The headline range is roughly 18 to 25% of order value as a base commission, with another 5 to 10% layered on for promoted listings, ad placements, and prime-time visibility. A restaurant doing serious volume on a top-tier aggregator placement can effectively be giving up 25 to 35% of every order to platform fees.
For a dish with a 65% gross margin (a reasonable benchmark for most Indian casual dining), a 30% commission cut means the actual margin retained on an aggregator order is closer to 35%. For pizza, biryani, and other high-volume staples where competition has compressed pricing, the number is often worse.
Now compare that to the direct-ordering alternative. A restaurant that captures the order on its own channel (a QR menu, WhatsApp, or a direct-call) and uses Porter for delivery pays a logistics fee that is typically a flat or distance-based charge, not a percentage of order value. On a ₹600 order, that is roughly ₹50 to ₹80 to Porter, versus ₹120 to ₹180 to an aggregator. The restaurant keeps the difference.
Multiply that across 30 to 50 delivery orders a day and the operator is recovering ₹2,000 to ₹4,000 of margin daily that was previously walking out the door as platform fees. That number is not a hypothesis. It is what restaurants doing this in Bangalore, Chennai, and Mumbai are seeing in their own books.
Three forces are pushing this shift in 2026 specifically.
First, aggregator commissions have crossed an invisible threshold. For years, operators absorbed the platform fee because the alternative (building a direct channel) was operationally complex. The commission climb, combined with the rest of the cost pressure on the P&L, has flipped the cost-benefit. Building a direct channel is now cheaper than continuing to pay the platform tax.
Second, third-party logistics has matured. Porter, Dunzo, and similar logistics-only partners give restaurants on-demand delivery without requiring exclusivity, without taking a percentage of order value, and without competing for the customer relationship. The logistics layer is now decoupled from the ordering layer in a way it was not five years ago. That decoupling is the structural change that makes aggregator escape viable.
Third, the QR menu has become the customer-acquisition surface that aggregators used to be. A guest who scans a QR menu in-store, places an order, and pays through that channel is now a known customer. The next time they want to order from that restaurant, they go direct. Every dine-in QR scan is a future direct-order pipeline if the operator is set up to capture it.
The combination is what makes this shift different from past attempts to "go direct." Earlier attempts failed because the logistics layer was missing, the customer acquisition was hard, and the operational overhead of running a parallel channel was punishing. None of those constraints are binding anymore.
The framing matters. Restaurants that go fully off the aggregators almost always lose money, especially in the first 12 months. Aggregators still drive meaningful new-customer acquisition, particularly in dense urban markets where discovery is dominated by app-based search. Walking away from that volume is operationally reckless.
The smart play is channel-mix optimization, not channel switching. The operator's job is to figure out which orders should run through which channel:
The aggregator channel is for new-customer acquisition, peak-hour overflow, and discovery in neighborhoods where the restaurant is not yet well-known. Pay the commission, treat it as a marketing line item, accept the lower margin in exchange for the funnel.
The direct channel is for repeat customers, dine-in conversions to delivery, and any guest who has already been through the door once. These are the orders where the restaurant is paying the aggregator a 25 to 30% commission to deliver a customer the restaurant already owns. That is the margin leak worth fixing.
The math is straightforward. If 40% of a restaurant's aggregator orders are repeat customers (a reasonable baseline for most casual dining), shifting half of those to a direct channel recovers somewhere between ₹50,000 and ₹150,000 a month in retained margin, depending on volume. That is the number worth chasing, and it does not require burning the aggregator relationship.
Here is the working model for Indian operators who want to build a direct-channel without losing aggregator volume.
Most restaurants think of QR menus as a dine-in convenience. The strategic version is different: the QR menu is the customer-acquisition surface that converts a one-time dine-in guest into a repeat direct-order customer. Every guest who scans the QR is a future direct-channel order if the system captures their phone number, their order history, and their preferences. Menuthere is built specifically for this loop: dine-in QR scanning, order capture, customer database building, and direct-ordering re-engagement, all from one menu surface.
The single biggest reason direct ordering fails is friction. If ordering through the aggregator takes 4 taps and ordering through the restaurant's own channel takes 12, the customer goes back to the aggregator no matter how much the restaurant promotes the discount. The direct channel has to match or beat aggregator UX. A QR menu with order, pay, and re-order built in clears this bar. A WhatsApp-only flow does not, except for the most loyal regulars.
The trap operators fall into is hiring delivery staff to "save on commissions." That replaces a variable cost with a fixed one and almost always loses money below 50 daily delivery orders. Porter, Dunzo, and Shadowfax give the restaurant on-demand delivery at variable cost, with no exclusivity, no minimum volume, and no headcount risk. For most independents, third-party logistics on demand is the right structure.
The operator's instinct is often to push every customer to the direct channel through aggressive in-store messaging. That backfires. The right play is gentler: a 10% direct-order discount, a "skip the platform fees" message printed on the bill, a WhatsApp message after the third aggregator order suggesting the direct alternative. The regulars who care about value will switch. The infrequent customers will stay on the aggregator. That is the desired outcome.
Most operators look at total monthly margin and feel okay. The hidden problem is that the aggregator orders are dragging down a blended number that the direct orders should be lifting. Tracking margin by channel (aggregator vs. direct) makes the leak visible and gives the operator something concrete to optimize. Without that visibility, the case for shifting channel mix is invisible to the people running the P&L.
The strongest moment to push direct-ordering adoption is right when food costs are spiking and the restaurant is being forced to raise prices anyway. The narrative writes itself: "Order direct and we can hold prices." That is a customer-friendly framing that turns a margin defense into a value story. Pushing the same campaign in a stable cost environment feels promotional. Pushing it in an inflationary environment feels honest.
Aggregator commissions in India have been climbing steadily for years and have now crossed the point where the aggregator-only channel mix is structurally incompatible with healthy margin. The operators who are quietly rebuilding their order flow, putting the QR menu at the center, using Porter for delivery, and converting their own regulars to a direct channel, are recovering 5 to 12 points of margin without losing aggregator volume.
This is not a Zomato story or a Swiggy story. The aggregators are doing what aggregators do. The story is about what restaurants can do back. The economic logic of the platform-only channel mix has shifted, the alternatives have matured, and the operators who notice early get to compound the margin recovery for years before the rest of the market catches up.
The framework is the same as any other channel-mix question in any other industry: pay for what the platform actually does (acquisition), do not pay for what the restaurant can do itself (retention). The tools to act on that distinction finally exist for independent Indian restaurants in 2026. The operators who use them are the ones whose 2026 P&Ls are quietly going to look very different from their 2025 P&Ls.
Build the direct channel without losing the aggregator volume.
Menuthere is a digital QR menu built for Indian restaurants that turns dine-in scans into direct-ordering customers, integrates with Porter for delivery, and gives operators full visibility into channel-level margin.
Sources: Industry commission rate data from operator interviews and trade publication reporting, Porter and Dunzo public pricing pages, restaurant operator P&L benchmarks, NRAI 2025 India Food Services Report.
Indian customers are skipping Zomato and Swiggy, ordering directly from restaurants and using Porter or Rapido for pickup. Here is the operator playbook.
Menu strategy
Around one third of diners order the very first item they see in a category.
Aggregators
Aggregator money arrives in 7 to 14 days, minus a third, in a lump you cannot reconcile. The vegetable vendor wanted cash on Monday morning.