You run a restaurant, the delivery orders keep dropping in through Glovo, Pyszne and Uber Eats, and on paper it looks like business is good. Then you count the month and see that on every delivered order the aggregator quietly took 20-30%, and on the cheaper dishes you basically cooked for free. Restaurant online ordering with no commission isn't a fantasy - it's your own ordering system or app, with online payments and an order landing straight in the kitchen, where you keep the full margin and own the customer instead of renting them from a platform.
What the aggregators actually take
Glovo, Pyszne.pl and Uber Eats don't run a charity. The standard commission on a restaurant order sits around 25-30% of the order value, sometimes a little lower if you negotiate volume, often higher once you add the "extras" the platform pushes.
Put real numbers on it. A customer pays 80 zł for an order. At 28% commission the platform keeps about 22 zł before you've paid for a single gram of food. Now subtract food cost (say 30%, another 24 zł), the driver if you deliver yourself, the packaging, the staff time. On a 80 zł order you're often left with single-digit złoty - and on a 35 zł pizza-for-one you can finish in the red. Run 600 delivery orders a month at an average ticket of 70 zł and 27% commission, and you've handed the aggregator roughly 11,000 zł in a single month. Over a year that's well past 130,000 zł.
But the commission isn't even the worst part. The worst part is invisible: the person who ordered from you is not your customer, they're Glovo's customer. You don't have their phone number, you can't text them a Friday promo, you can't bring them back for free. Next week the app shows your kitchen next to four cheaper rivals, and you pay commission again for a guest who already loves your food.
What you don't own on the aggregators
On the platforms you build turnover, not a business that's worth anything when you sell it. Three things stay on their side of the wall.
- The customer data. Names, phones, addresses, order history - all of it lives in the aggregator's account, not yours. No loyalty SMS, no "we miss you" coupon, no birthday offer.
- The relationship. The guest remembers "I ordered on Pyszne," not the name of your kitchen. You can't build loyalty on a logo that isn't yours.
- The margin on repeat orders. A regular who orders from you every Friday pays you commission every single Friday. On your own channel that same regular costs you almost nothing after the first order.
Your own ordering flips the math. The first order through your own system might cost a bit in ads, but the fifth, tenth and fiftieth from the same regular are nearly free - because you already hold their number and your own channel.
What an own ordering system includes
I don't build a toy "order form." I build the real machine that replaces the aggregator for your own traffic: a clean digital menu with photos, categories and add-ons (extra cheese, sauces, sizes), a cart and checkout that works on a phone in a few taps, BLIK and online payment through Przelewy24 or PayU so the money is in before the food leaves, delivery zones with a minimum order and a fee per area (or pickup), and - the part that actually matters in a busy kitchen - the order pings the kitchen instantly, either to a Telegram channel the cooks watch or straight to a receipt printer, with no one refreshing a browser tab.
On top of that: opening hours and "kitchen closed" logic, an estimated delivery time, promo codes for your own loyalty, and a simple admin where you flip a dish to "sold out" in two seconds. It can be a fast mobile-friendly web ordering page, or a full native app your regulars install once and reorder from in three taps - I'll get to which one fits you below. Either way I build it as a proper product, the same way I built Body Forge, a native app with payments and clean ordering flows, not a patched-together plugin.
Web ordering vs an app - which one you actually need
Both keep the margin. They differ in who you're keeping it from.
A web ordering page is the right start for almost everyone: nothing to install, it works from a Google or Instagram link or a QR code on the table and the flyer, and it indexes for searches like "pizza delivery [your district]." It's cheaper and faster to ship.
A native app earns its place when you have a real base of regulars and you want them to reorder in three taps, get push notifications about the Friday deal (push you don't pay 28% for), and collect loyalty stamps. The honest order is: launch the web ordering first, and add the app once your repeat orders justify it. I build the app side as mobile apps - your own ordering app for iOS and Android, wired to the same menu, payments and kitchen notifications.
What it costs and when it pays off
Numbers instead of theory. A mobile-friendly web ordering system with BLIK/online payment, delivery zones and kitchen notifications runs from about €1,500-2,500 depending on the menu size and integrations. A full native ordering app for iOS and Android starts from €6,000-12,000. Hosting and domain are roughly €60-90 a year, plus the payment processor's small per-transaction fee (a fraction of a percent plus pennies, nothing like 28%).
Now the payback. Say the aggregators take 11,000 zł from you a month. You don't need to move everything - just your regulars. Shift even 30% of delivery orders to your own channel and you stop paying commission on that slice: roughly 3,300 zł saved every month, every month after, plus you finally own those guests' numbers. At that pace a web ordering system pays for itself in about three to four months, and from then on it's pure margin you keep plus a customer base that's yours forever.
The realistic point: keep the aggregators running for discovery (new people who don't know you yet), but push every repeat guest to your own ordering with a flyer in the bag - "next order 15% off, order direct at [yoursite]." Aggregators stay your acquisition channel, your system takes the regulars. That's where it pays off - not by switching Glovo off overnight, but by quietly moving the people who already love your food onto a channel that doesn't tax every plate.
The timeline is predictable: I build and launch a web ordering system in 2-3 weeks, a native app in a few weeks more, and the first direct orders usually land in the first week if we put a QR code on the table and the delivery bag.
If the aggregators are eating a third of every order and the customers still aren't yours - I'll build you your own ordering system or app with BLIK, delivery zones and an order that pings the kitchen the second it's paid. Write to me on the contact page and I'll work out, on your real order volume, when it pays for itself - quote in 24 h, free.
FAQ
How much commission do Glovo, Pyszne and Uber Eats take? Usually 25-30% of the order value, sometimes a bit less with negotiated volume, often more once paid promotion is added. On an 80 zł order that's around 22 zł gone before food cost. The bigger hidden cost is that the customer stays the platform's, not yours - no contact, no way to bring them back for free.
What does my own restaurant ordering system include? A digital menu with photos and add-ons, a phone-friendly cart and checkout, BLIK and online payment via Przelewy24 or PayU, delivery zones with minimum order and fee per area (or pickup), and instant kitchen notifications to Telegram or a receipt printer. Plus opening-hours logic, delivery-time estimates, promo codes and a simple "sold out" admin.
Web ordering or a mobile app - which is better for a restaurant? Start with a web ordering page: nothing to install, works from a link or QR code, indexes in Google, cheaper and faster. Add a native app once you have enough regulars who'd reorder in three taps and want push promos and loyalty - that's where the app earns back. I usually launch web first, then the app when repeat orders justify it.
How much does a restaurant ordering system cost and when does it pay off? A web ordering system runs from about €1,500-2,500; a native iOS/Android app from €6,000-12,000. If aggregators take around 11,000 zł a month and you move even 30% of orders to your own channel, you save roughly 3,300 zł a month, so a web system typically pays for itself in three to four months - after that it's margin you keep plus customer data that's yours.
Do I have to drop the aggregators completely? No, and you shouldn't at first. Keep Glovo and Pyszne for discovery - new customers who don't know you yet - and push every repeat guest to your own ordering with a flyer and a direct-order discount. The aggregators become an acquisition channel while your system takes the regulars, where the real margin lives.
