The refusal rate is the number that decides this, not the convenience
Across Kuwait fashion brands we work with, open cash on delivery lands somewhere between 12% and 25% refusal at the door. Prepaid orders on the same store refuse at 2% to 5%. That gap is the whole argument. A refused COD order still costs you the delivery fee both ways, usually 2–3 KD, plus the driver's second attempt, the restocking labour, and a piece that comes back creased or smelling of perfume. On a 25 KD dress at 55% margin, three refusals wipe out the profit on four good orders.
The refusals are not random. They cluster on impulse buys from Snapchat and Instagram stories, on orders placed after midnight, on first-time buyers with no history, and on anything the customer expects to try on before paying. Drivers in Kuwait get asked to wait at the door while she tries the piece far more often than anyone admits, and once trying is allowed the refusal rate on that order roughly doubles. Ask your courier for a refusal report split by governorate and by hour. Most brands find Jahra, Farwaniya and post-1am orders carry double the average.
Before you change anything, make sure you can actually see the number. If your orders live in WhatsApp threads and an Instagram DM folder, you do not have a refusal rate, you have a feeling. A proper store on Shopify with KNET and Tabby wired into checkout puts payment method against fulfilment status in one report, so you can price COD honestly instead of guessing at it. Give it four clean weeks of data before you touch anything in the checkout.
Ready to start your Shopify store?
Start a free trial and try the platform for yourself.
COD still earns its place on the first order and under 40 KD
Do not kill COD outright. In Kuwait it is still the trust bridge for a customer who has never heard of you. On cold Snapchat or TikTok traffic, offering COD typically converts 20% to 40% better than prepaid-only, and that first delivered box is what buys you the second order at full prepaid. So keep it for order one. The honest test is repeat behaviour: once a customer has two delivered orders behind her she does not need COD any more, and leaving it on simply hands her an easy way to change her mind at the door.
Price is the other axis, and it runs opposite to what most owners assume. Refusal climbs with ticket size. Under 15 KD people take the box because arguing is not worth the hassle. Above 60 KD the customer starts doing the maths at the door, and that is where you lose whole orders. So cap COD instead of banning it: allow it up to roughly 40 KD, charge 1–1.5 KD for the privilege, and make that fee vanish the moment she pays by KNET. Above the cap, offer Tabby instead of cash.
Watch the categories too. Abayas, jalabiyas and anything sized tightly refuse far more than scarves, bags and accessories. Ramadan and the weeks before Eid push COD volume up and refusal up with it, because everyone is ordering three sizes to keep one. During Jul and Aug, when half your list is travelling, COD refusals spike again on addresses nobody is standing at. Turn the cap down to 25 KD in those windows. It costs you a few orders and saves you a pile of returned stock.
Make KNET and Tabby the easier choice, not the punished one
The shift works when prepaid is easier, not when cash is punished. Put KNET first in the checkout, above cards, with the logo visible, because that is the button Kuwaiti customers trust without thinking. Make the COD fee explicit at 1–1.5 KD and show the KNET total right next to it, so the saving is visible at the moment of choosing. Most brands move 15 to 25 points of their COD share within two months on that change alone. Do not announce it as a policy on stories. Change the checkout quietly and let the price difference talk.
Tabby does the heavy lifting on exactly the pieces that used to need cash. A 90 KD abaya split into four payments of 22.5 KD feels smaller than the same amount handed over at the door, and you are paid in full upfront while Tabby carries the default risk. Their fee, usually somewhere between 4% and 7% depending on your volume, is far cheaper than a 20% refusal rate on the same basket. Switch Tabby on above 25 KD and show the per-instalment price on the product page, not only at checkout. That is where it changes the decision.
Then remove the reasons people refuse in the first place. Publish real measurements per piece in centimetres, not S, M, L. Shoot on a body that looks like your customer, not a studio mannequin. State the return window plainly, 7 days unworn for exchange or store credit, so trying at the door feels unnecessary. Send one WhatsApp confirmation an hour after the order and one before dispatch. Brands that do all four usually watch COD refusal fall from the low twenties to under 10% without removing cash on delivery at all.