The aggregator owns your customer unless you take them back
Talabat and Deliveroo are not your customer list. You never see the phone number, you cannot message the person who ordered from you eleven times, and the 25–30% commission is rent you pay again on every single reorder. That is the design of the platform, not a flaw in it. So treat every delivery bag as your one chance to move that person to a channel you control. A printed card, a QR code on the box, a sticker on the lid — anything that gets a WhatsApp message started costs you fils and pays back for years.
The card needs a reason, not a logo. 'Order direct on WhatsApp and get 15% off' works. 'Follow us on Instagram' does not. Give the discount on the direct channel only, and keep giving it — the moment you match the app's price you have removed the reason to switch. If you are taking direct orders seriously, run them through your own ordering page with KNET checkout rather than a staff member typing addresses into WhatsApp at 9pm. Most restaurants we work with move 15–30% of delivery volume direct within six months.
Then actually use the list. Two messages a month, not eight. One is worth opening — the Thursday family platter, the Ramadan pre-order sheet, the new item with a photo you took yourself. The other is timed to that one person: they ordered every Friday for two months, then went quiet for three weeks, so they get a single message. Take consent when you take the number, keep the opt-out honest, and send between 6pm and 9pm. A 2,000-number WhatsApp list in Kuwait outperforms 40,000 Instagram followers, every single time.
Ready to start your Shopify store?
Start a free trial and try the platform for yourself.
Loyalty for food is a punch card, not a points economy
Points that convert at some invisible rate are a coffee-chain product. A restaurant doing 60 orders a day needs something the cashier can explain in four seconds. Buy nine, the tenth is on us. Spend 40 KD in a month, the next delivery is free. That is the whole scheme. The reward has to be food, not a discount on food — a free item feels generous and costs you the food cost, roughly 30% of what the customer values it at. A 5% discount costs you 5% of your margin and nobody remembers receiving it.
Do not build an app. Nobody in Kuwait is installing a fourth restaurant app for a shawarma place in Salmiya. The phone number is the card. Whoever takes the order types the number in, the system counts, and the reward fires automatically over WhatsApp. Off-the-shelf loyalty tools that sit on top of your POS run 20–60 KD a month here; a custom app quoted at 2,500 KD is money set on fire. Start with a paper card at the counter if you have to — the maths of retention does not care how the card looks.
Measure one number: what share of this month's customers also ordered last month. For delivery in Kuwait a healthy repeat rate lands between 25% and 40%, and below 20% you have a product or consistency problem that no loyalty scheme will paper over. Compare each month against the same month last year, not against last month — Ramadan, the Jul–Aug travel exodus and Hala February all move the number for reasons that have nothing to do with you. Judge the programme after 90 days, not after 30.
One extra visit a month is the whole business
Here is the uncomfortable part. The biggest reason a regular stops coming is not a competitor's offer — it is the one order that arrived cold, short a sauce, or with half the portion because the good cook was off that night. People do not complain, they just stop. So the retention work is boring: weigh the portions, photograph the plate you want and stick it above the pass, and taste your own food out of a delivery bag after 25 minutes in a car in July. Fix that before you spend a single fils on marketing.
Now the maths, because the target is smaller than it sounds. Say you have 300 regulars and the average ticket is 4 KD. Move each of them from three visits a month to four and you have added 1,200 KD a month, 14,400 KD a year, with zero new customers and no ad budget. Buying 300 new customers of that same value would cost you 3–8 KD each in ads plus the discount it takes to get them through the door the first time. The list you already have is the cheapest growth available to you.
To actually earn that extra visit, work the calendar and the gap. Fri–Sat family orders, Thursday night, exam week, the first week after payday. Then the lapse trigger: anyone who ordered twice and has gone quiet for 21 days gets one message with a real reason to come back, not 'we miss you'. Expect 8–15% of that lapsed list to order within a week, which is good enough to run every month forever. Track the result, keep the message that won, delete the rest, and stop rewriting your Instagram bio.