Do the commission maths on one 8 KD order
Take a typical order: 8 KD. The app takes 22–30% depending on your contract and whether you use their drivers or your own — call it 25%, so 2 KD. Food cost on a decent menu runs 30–35%, another 2.400–2.800 KD. Packaging that survives a twenty-minute run to Jahra costs 0.250–0.400 KD, not the 0.100 KD box you use in the dining room. The app absorbs the KNET and card fees, but you are paying for them inside the commission. You are left with roughly 2.500–3 KD before rent, salaries and the marketing that brought the customer in.
Now run the same 8 KD through your own counter. No commission, cheaper packaging, and the contribution is close to 5 KD. That gap is the whole argument. It gets worse once you join the app's discount campaigns, because a 25% off promo is usually funded by you and not by them: the 8 KD order becomes 6 KD of revenue, and the commission may still be calculated on the pre-discount value depending on the clause. Read that clause. Plenty of Kuwait restaurants discover in Ramadan that they are selling a negative-margin item at volume and calling it growth.
Before you rage-quit, price what the commission actually buys. Talabat and Deliveroo put you in front of people in Salmiya and Hawally who have never heard of you, and they handle the driver, the refund, the map pin and the payment. Acquiring that same customer yourself through Snapchat or Instagram costs most Kuwait restaurants 1.500–4 KD for a first order. If commission is 2 KD an order and your own acquisition is 3 KD, the app is cheap — until that customer becomes a repeat buyer. The whole game is moving repeat customers off the app, not new ones.
Ready to start your Shopify store?
Start a free trial and try the platform for yourself.
Your ranking inside the app is earned in the kitchen
Ranking inside Talabat, Deliveroo and Jahez is mostly operational. The sort is driven by how many people who see you actually order, your acceptance rate, your quoted prep time versus the real one, cancellations and out-of-stock items, your rating over the last 30–90 days, and how much you take part in their campaigns. Distance and zone matter too, because you rank inside a delivery polygon and not across Kuwait. A restaurant in Kuwait City that takes nine minutes to accept and marks four items unavailable every Thursday will sit below a worse restaurant with a tighter kitchen.
So the fixes are unglamorous. Accept in under 60 seconds. Set an honest prep time: 25 minutes you hit beats 15 minutes you miss. Switch off an item at 9pm when it runs out instead of cancelling the order at 9:40. Reshoot your top eight dishes, because the photo drives the tap-through that feeds the ranking. Keep the menu tight, since 90 items on an app is a prep-time problem rather than a choice. And treat the first two to four weeks after launch as the new-store boost — push hardest then, because that visibility never comes free again.
Paid placement inside the apps works, but budget it as media, not as rent. Most Kuwait restaurants that run it well spend 150–400 KD a month on in-app promotion, and expect orders to rise while it runs and fall when it stops. Free delivery above a 5–6 KD basket is usually the highest-return lever you have, because it raises average ticket at the same time. What does not work is a permanent 30% discount. You train the customer to order only at 30% off, and a Kuwaiti customer compares three apps in the same minute.
Own drivers, a fleet, or the apps — most restaurants need all three
A delivery driver in Kuwait costs 180–250 KD in salary, and once you add the bike, fuel, insurance and residency you are near 300–350 KD a month all-in. A busy driver does 15–25 drops a day, roughly 450–600 orders a month, which puts your cost per drop at 0.500–0.750 KD. A third-party fleet charges most restaurants 0.900–1.500 KD a drop with no headcount risk and no idle driver in July when half of Kuwait is travelling. Both are far below the 2 KD you hand the app on an 8 KD order. The catch is that you have to supply the demand yourself.
That is why the direct channel is a build, not a switch. Start with what already exists: WhatsApp orders from your regulars, a QR code on the table and on the bag, and one simple ordering page. Most Kuwait restaurants get there fastest with their own Shopify store wired to KNET and a fleet doing the driving, because you keep the customer's number, the order history and the margin. Budget 400–900 KD to set it up properly and 40–60 KD a month to run it. Then spend the commission you saved on bringing people to it.
The realistic target is not zero apps. Most Kuwait restaurants start at 90% aggregator and should aim for 60–75% aggregator and 25–40% direct within 12–18 months. Pushing past that usually means refusing orders you could have taken. Keep prices identical on the app and direct, because Kuwaiti customers check and a cheaper direct price breaches most aggregator contracts anyway. Compete on what the app cannot copy: a free item on the fifth direct order, a WhatsApp message the night before Eid, and a driver who knows the building in Salmiya without calling.