Own delivery app or iFood: when to build yours

Every restaurant owner who sells on iFood has done this math at the end of the month. Add up the orders, check the payout and ask where the difference went. Part of it went to commission. Part went to the online payment fee. And the customer who ordered usually doesn't even know your restaurant's name: they know they ordered "on iFood."

That's where the question comes from: is it worth having your own delivery app? The answer isn't yes or no. It depends on how much you sell, how many customers come back and how much of your margin already goes to the marketplace. In this article I show the numbers behind that decision and when each path makes sense.

What iFood charges restaurants in 2026

According to iFood's official plans page for restaurants, checked in October 2026, there are two main plans:

• Basic Plan: 12% commission on delivery orders, a 3.2% fee on orders paid through iFood and a R$110 monthly fee. The store handles delivery itself.
• Delivery Plan: 23% commission, the same 3.2% fee on online payments and a R$150 monthly fee. Delivery is handled by iFood's partner couriers, with tracking and support until the customer gets the order.

The monthly fee only applies to stores that bill more than R$1,800 a month on the platform, and the Delivery Plan's availability depends on the region. iFood notes that rates may vary, and the number that applies to you is the one in your Partner Portal.

In practice, a restaurant on the Delivery Plan that gets most orders paid online hands the marketplace close to a quarter of each order. In a month with R$50k in orders, 23% commission comes to R$11,500. Add 3.2% for online payment and that's another R$1,600. Plus the monthly fee.

Where Brazilian restaurants sell today

The marketplace isn't the villain. It's the channel that sells the most. An Abrasel survey published in March 2025, with 2,176 business owners in the sector, shows how delivery revenue splits:

• 54% comes from marketplaces;
• 26% comes from WhatsApp;
• 12% comes from their own app or website;
• 8% comes from phone orders.

The same survey shows that 78% of businesses with delivery are on marketplaces, 63% use WhatsApp and 39% invest in their own app or website. The number that stands out most is WhatsApp's. Abrasel's president, Paulo Solmucci, put it this way: "The growth of WhatsApp is natural, because it gives businesses more control."

Control is the word. An order that comes in through WhatsApp pays no commission and leaves the customer's contact with the restaurant. The problem is that WhatsApp wasn't built to take orders. The attendant types out the menu, checks the address, works out change and makes mistakes when the kitchen is slammed. Your own app, or a well-built ordering website, does that same job without an attendant in the middle.

When your own delivery app pays off

Your own app pays for itself when three things happen at once.

1. You have customers who come back. An app is a repeat-purchase tool. People who order every week install it, save their card and order in two taps. People who order once a year won't download anything. If you don't know how much of your revenue comes from repeat customers, that's the first number to run.

2. Commission already weighs on your margin. Once marketplace commission starts setting your menu prices, your own channel stops being a luxury. Every order that moves from iFood to your app puts the commission share back in your till.

3. You already handle delivery. If you're on the Basic Plan, you already have couriers. In that case, your own app only changes where the order comes in. If you rely on the marketplace's logistics, you need to solve delivery first. The Abrasel survey shows that 39% of businesses use their own couriers, 30% hire logistics companies and 26% use on-demand couriers.

A common question: does your own app pay a fee to Apple and Google? Not on orders. App Store guideline 3.1.3(e) says purchases of physical goods or services consumed outside the app must use another payment method, such as a card. Food falls under this rule. I broke down the store accounts and fees in what the App Store and Google Play charge in Brazil.

When the marketplace is still better

In some cases, your own app is money spent too early:

• New restaurant or new neighborhood. Nobody searches for your name yet. The marketplace is where customers search for "pizza near me." First you need to be found.
• Infrequent orders. If customers order once every two months, they won't keep your app on their phone.
• Low volume. If your monthly commission is still small, the cost of building and maintaining an app takes a long time to pay back.
• No way to promote it. Your own app doesn't bring in customers on its own. Someone has to tell the people who already buy: a coupon in the bag, a WhatsApp message, a sign at the counter.

In these cases, a fast ordering website connected to WhatsApp is usually the first step. I explained the difference between formats in native, hybrid or PWA: which to choose.

What your own delivery app needs

If the numbers work, the app has to compete with the experience customers already know. The minimum:

1. A menu that's easy to update, with photos, add-ons and sold-out items that disappear instantly.
2. In-app payment with Pix and a saved card, so the second order takes seconds.
3. Saved address and delivery fee by area, calculated without the customer having to ask.
4. Order tracking with notifications: received, being prepared, out for delivery.
5. Simple loyalty: cashback, points or a free item every so many orders. It's the reason for customers to switch from the marketplace to your app.
6. A restaurant dashboard with orders arriving in one place, ideally connected to the system you already use at the register.
7. Customer data in your name, so you can message anyone who hasn't ordered in 30 days.

What changes the price of this project is how much of this goes into the first version and how many integrations it needs. I explained what drives the budget in how much it costs to build an app in 2026.

What this means for you

The decision isn't "iFood or your own app." It's almost always "iFood and your own app." The marketplace stays as the storefront for new customers. Your own app keeps the people who already know your food and order again. Every order that switches channels keeps the commission in your till and the customer's contact with you.

Before asking for a quote, run three numbers from your Partner Portal: how much you paid in commission over the last six months, how many customers ordered more than once in that period and how much your delivery costs today. With those numbers, you can tell whether the app pays for itself and how fast.

At the Studio, we set a fixed price after discovery and put the deadline in the contract. The code, the store accounts and the customer data are in the restaurant's name, not ours. If the numbers work for you, see how we plan and ship apps for businesses that sell every day.