How to reduce failed cash on delivery orders in the UAE

A failed cash on delivery order costs you the delivery twice and the sale once. Quiqup, a UAE delivery company, says a returned parcel means paying for logistics twice, and in May 2026 it put return-to-origin rates for many UAE merchants at between 20% and 35%, with under 10% a realistic target.

Quiqup's range is a delivery company's own estimate for returned parcels in general, not an official figure and not a cash-only number. The post on what RTO means explains the measure. Start by finding your own rate, then put a price on it.

What does one failed order cost?

A failed order costs every dirham that leaves the business when the parcel comes back, plus the ad spend that won the order and the margin you never earned. The table adds up the part you can count from invoices.

Say you ship 1,000 cash on delivery orders a month and 25% fail, a made-up rate inside Quiqup's range. That would be 250 parcels and AED 12,000 of cash lost every month. At 10% the loss would be AED 4,800, which leaves AED 7,200 a month as the most that prevention can save on cash alone, or about AED 7 for each order.

Orders that turn from failed into delivered are worth more than that, because they also earn their margin. The ad dashboard hides part of the damage as well. It counts an order when it's placed, so say you pay AED 40 per placed order and 25% fail, which makes each delivered order cost about AED 53 in ads. The post on what you really keep from ROAS shows how to correct for it, and Magnetic Keys' e-commerce marketing optimises against contribution margin rather than platform ROAS.

How do you cut failed cash on delivery orders, step by step?

Work through these seven steps in order, because the first one tells you which of the others will pay.

1. 01Split your failures by cause. Export the last 100 failed orders and tag each one as refused at the door, customer unreachable, wrong or incomplete address, nobody home, or cancelled after dispatch. If your courier doesn't give reasons, ask for its attempt log, and if it has none, tag orders by phone for two weeks.
2. 02Confirm every cash on delivery order on WhatsApp before it ships. If the customer doesn't answer, send one reminder, call once, and don't ship an order nobody has confirmed.
3. 03Check the address and phone at checkout, not at the door. If an order still arrives without a map pin or a working mobile, such as one taken over Instagram DMs, the confirmation message asks for them before anything is packed.
4. 04Make paying online the easy choice. If failures don't fall in the group that was offered a nudge, drop the nudge and try another.
5. 05Ask the customer for a delivery day and window. If your courier can't keep windows, say so when you ask instead of promising what you can't deliver.
6. 06Chase every failed attempt within the hour: call, message, rebook. If the customer refuses, record the reason in one tap, because that reason feeds the next round of tagging.
7. 07Review weekly by cause, area, courier and basket size. If one courier or one area dominates the failures, change the routing before you change the checkout.

How should the WhatsApp confirmation work?

Quiqup recommends WhatsApp order confirmation, prepaid nudges and address validation, and the confirmation is the one to build first. WhatsApp is where UAE buyers already are: in GWI's Q2 2026 survey, published by DataReportal, 84.1% of UAE internet users aged 16 and over said they had used it in the past month, ahead of YouTube, Instagram, TikTok and Facebook.

Send the message within minutes of the order, in the language the customer used at checkout, and keep it to the order. A workable message is short: "Hi [name], your order [number] from [shop] is ready to ship. Amount to pay on delivery: AED [amount]. Reply 1 to confirm, or send your location pin if the address needs changing. Delivery: [day], [window]." Write the Arabic version natively rather than translating the English, and add a line at checkout saying you'll confirm on WhatsApp.

Decide the no-reply rule before launch. Start with two hours inside working hours, one reminder, one call, then hold the order, and adjust from your own data. Expect to cancel some orders that would have been delivered, because a few real customers won't answer in time.

Judge every fix on profit per order placed, not on failure rate alone.

Meta charges for WhatsApp Business Platform messages per delivered message, a model that replaced conversation-based pricing on 1 July 2025. From 1 October 2026 Meta charges for customer service messages again, with each business number getting 1,000 free delivered service messages a month, so any guide that calls service replies free is out of date. Rates differ by market and message category, and Meta's change table also shows a marketing rate increase for the UAE on that date.

Price the flow before you build it. A flow with a confirmation, a reminder and an out-for-delivery notice sends up to three messages per order, so multiply three by the current UAE rate for the category each message falls under on Meta's rate card, then add the staff time spent on reminder calls. Compare the total with your prevention ceiling, which is the cash a failed parcel costs multiplied by the failures you expect to stop. The post on how the WhatsApp Business API works covers the set-up.

Collect a map pin, a building name and a mobile that answers

Collect a map pin, a building or villa name and a mobile number that can receive WhatsApp, and don't accept an order without all three. Address validation, which Quiqup recommends, means little unless the checkout refuses an order that lacks them.

In Dubai, add the Makani number if the customer has one. Where the pin and the typed address disagree, ask which is right in the confirmation message rather than guessing.

Treat the confirmation as the phone check too. A message that never delivers means the number is wrong or isn't on WhatsApp, which is cheaper to learn before dispatch than at the door, and Meta doesn't charge for messages that aren't delivered.

Nudge towards prepaying, but measure what the nudge costs

Prepaying is where the market is heading, but the pace is yours to find out. Checkout.com's State of Digital Commerce in MENA 2025 report says cash on delivery usage in the UAE has fallen 53% since 2020, in a survey run with YouGov across 16 countries. That's a fall in usage, not today's share of orders, and no reliable current share was available when this guide was written, so read your own order export.

Incentives are the expensive lever. Say 30% of buyers take an AED 10 prepay discount across 1,000 orders, which comes to AED 3,000 a month and about 42% of a prevention ceiling of AED 7,200. Most of that money goes to buyers who'd have accepted the parcel anyway, because a customer who might refuse it is also less likely to pay for it upfront.

Cheaper nudges come first. Make card the default choice, label cash on delivery as "pay at the door", and offer pay later on bigger baskets. Then run any discount on half your traffic for two weeks and compare profit per order placed across the halves, discount included.

Restrict before you remove. If cash on delivery still loses money after confirmation and address checks, require prepayment above a basket value or in the areas where failures cluster, and keep the option open elsewhere.

Ask for the day and window instead of assuming them

Ask the customer for a day and a window instead of guessing, because the UAE working week isn't the same for everyone. The federal government's weekend has been Saturday and Sunday with a half day on Friday since 1 January 2022, Sharjah's government moved to a Friday to Sunday weekend on the same date, and private companies can set their own week.

Offer windows you can keep. A promised evening slot that the courier can't hold produces the same failed attempt you were trying to avoid, so ask your courier two things before you build the picker: whether it can hold a window, and whether the driver calls before arriving.

Sale periods raise the stakes, because the same failure rate means more failed parcels. Amazon Ads describes White Friday as beginning on the Friday after US Thanksgiving, which makes it 27 November in 2026 by the calendar, although retailers hadn't announced dates by 7 November 2026. Have the confirmation flow running before then, and read the White Friday plan for the margin side.

Start this week by tagging your last 100 failed orders by cause. An afternoon's work will show whether refusals, bad addresses, dead phone numbers or missed windows account for most of the damage, and that decides what to build first.