What one refused parcel really costs, why moving from 60% to 80% delivered can nearly triple profit, and nine fixes in the order they pay off.
Every COD seller in Pakistan learns the same lesson the hard way: an order is not a sale. A parcel that comes back costs you delivery, return charges and the advertising you paid to get it. This guide puts rupee numbers on that problem and walks through the fixes in the order they pay off — starting with the ten-second phone call most sellers skip.
Every refused COD parcel costs you delivery both ways plus the ad spend that brought the order. The fixes that pay off first are confirming every order by call or WhatsApp, showing the real product and total price in your ads, dispatching quickly and blocking repeat refusers. Moving from 60% to 80% delivered can nearly triple your profit.
Let’s take a normal product and normal numbers. Your own figures will differ, so swap them in as you read.
| Item | Example amount |
|---|---|
| Selling price (COD) | Rs 2,999 |
| Supplier cost | Rs 1,100 |
| Delivery charge (forward) | Rs 250 |
| Return charge if refused (varies a lot by courier) | Rs 150 |
| Advertising cost per order | Rs 600 |
When the parcel is delivered, you keep Rs 2,999 − 1,100 − 250 = Rs 1,649 before advertising. When it is refused, you get nothing and pay Rs 250 + Rs 150 = Rs 400 in courier charges. The Rs 600 you spent on the ad is gone either way.
Now look at 100 orders from the same ads (Rs 60,000 of ad spend) at different delivered rates:
| Delivered | Earned on delivered | Lost on refused | After ads |
|---|---|---|---|
| 60 of 100 | Rs 98,940 | − Rs 16,000 | Rs 22,940 |
| 70 of 100 | Rs 115,430 | − Rs 12,000 | Rs 43,430 |
| 80 of 100 | Rs 131,920 | − Rs 8,000 | Rs 63,920 |
| 90 of 100 | Rs 148,410 | − Rs 4,000 | Rs 84,410 |
Same product, same ads, same price. Moving from 60% to 80% delivered nearly triples the profit. In this example, every extra 10 points of delivered rate is worth about Rs 20,000 per 100 orders. That is why the habits below matter more than finding a slightly cheaper supplier.
The break-even point here is a delivered rate of about 49%. Below that, the more you sell, the more you lose.
When we look at refused orders, almost all of them fall into a handful of buckets:
Every fix below targets one or more of these.
This single habit removes most fake orders and many “changed my mind” orders. Call or send a WhatsApp message within a few hours of the order. A simple script works:
Assalam o Alaikum, main [store name] se baat kar raha hoon. Aap ne [product] order kiya hai, total Rs 2,999 cash on delivery, delivery charges shamil hain. Parcel 2 se 3 din mein aa jayega. Kya aap address confirm kar dein ge aur bata dein ke kis waqt ghar par hote hain?
If the customer does not answer after two tries on different times of day, do not ship. An unconfirmed COD order is a coin toss with your Rs 400.
Use a supplier with stock in Pakistan who dispatches in one or two working days. The difference between a parcel arriving on day 2 and day 6 shows up directly in refusals. If an order sits for more than three days for any reason, reconfirm it before it goes out.
Add fields to your checkout for area, nearest landmark and an alternate phone number. “House 12, Street 4” means little to a rider in a new area; “near Al-Fatah, opposite the petrol pump” gets delivered.
The ad, the product page and the confirmation call must all show the same total. If you charge delivery, say it upfront. “Free delivery” that turns into “plus Rs 250” at the door is a refused parcel.
Film the real product, at its real size. The ads that bring the cheapest orders are often the ones with the highest refusals, because they overpromise. Judge every ad by cost per delivered order, not cost per order.
When the courier marks “customer not available”, call the customer yourself within a few hours and arrange a new time. Many couriers re-attempt, but a call from the store saves a large share of these parcels.
Watch for repeated orders from the same number, names like “test” or “abc”, impossible addresses and several orders from one person for different names. Keep your own list of numbers that refused before, and ask them for a small advance next time.
For products above roughly Rs 5,000, some sellers ask for a small advance through Easypaisa, JazzCash or bank transfer — enough to cover delivery. You will lose some orders, but the ones you keep are far more likely to be accepted. Test it on one product before applying it everywhere.
A courier that is Rs 20 cheaper but delivers 5% fewer parcels is more expensive. Track delivered rate by courier and by city for a month, then decide. Many sellers use two couriers and send each city to the one that performs best there.
An overall delivered rate hides the problem. Once a week, break it down:
Check returns quickly. Make sure the item came back complete and undamaged, and confirm with your supplier how returned stock is handled and what you are charged. Many dropshipping suppliers return the item to stock but still charge the courier fees, so the Rs 400 in our example is real money.
Problem-solving products with an obvious use — a foot file, a car mug, a hot water tap — tend to have better delivered rates because the buyer knows exactly what they are getting. Browse examples in the Pakistan product list.
Many of our readers run stores in more than one market. The refusal problem looks different in each, and so does the fix.
Cash on delivery is common, and sellers there call refused parcels “RTO” just as Pakistani sellers do. The same habits work: confirm by phone or WhatsApp, check that the courier covers the customer’s PIN code before you accept the order, and give a small reason to prepay. A modest discount for paying by UPI turns a share of COD orders into prepaid ones, and prepaid orders are almost never refused.
COD is still a big part of online shopping in both countries. Addresses are the main extra risk: ask for a map pin on WhatsApp, and in Saudi Arabia ask for the customer’s National Address where possible. Instalment options such as Tabby and Tamara give buyers a way to pay upfront without paying everything at once. Our COD guide for the UAE and Saudi Arabia covers pricing and confirmation there in detail.
Here most orders are paid by card, so parcels are rarely refused at the door. The risk moves to returns and chargebacks instead:
| Market | What customers can usually do | What to plan for |
|---|---|---|
| UK and EU (Germany, France, Italy, Spain, Austria) | Cancel most online purchases within 14 days of delivery, without giving a reason | A clear returns page, a returns allowance in your margin, refunds within the legal time limit |
| United States | No single national returns law; customers expect a store returns policy, often around 30 days | Write your policy clearly; chargebacks cost a fee on top of the lost sale |
| Australia | Consumer law guarantees cover faulty or misdescribed goods; change-of-mind returns depend on your policy | Accurate descriptions; handle faulty items quickly |
| Canada | Rules vary by province; store policy matters most | Clear policy and fast replies to complaints |
These are general points, not legal advice — check the consumer rules for each country you sell to. The cost logic is the same as in Pakistan: every return costs you shipping, sometimes twice, plus the ad that brought the order. For cheap items shipped across borders, return postage can cost more than the product. Many sellers refund without asking for the item back in those cases and treat it as part of the returns allowance.
| Market type | Your biggest loss | Best first fix |
|---|---|---|
| COD (Pakistan, India, UAE, Saudi Arabia) | Refused parcels | Confirm every order before dispatch |
| Cards with a legal right to return (UK, EU) | Change-of-mind returns | Honest photos, real size, clear sizing and specs |
| Cards without a national returns law (US, Canada) | Returns and chargebacks | Clear policy, tracking numbers, fast customer replies |
There is no official national figure, and numbers shared in seller groups vary wildly by product and city. Work out your own break-even rate with the method above, then aim well above it. In our example, 49% was break-even and 80% made a healthy business.
It can reduce “is this real?” doubts, but it also makes it easy to refuse on a whim. If your product is good and matches the ad, it can help. Check what your courier allows and test it.
Yes. In the example above, each refused parcel costs Rs 400 plus the ad. A two-minute call that saves one refusal in five pays for itself many times over.
Delivered orders divided by dispatched orders for the same period, after all returns have come back. Measure each week’s orders once they are fully settled, not orders still in transit.
In India, the UAE and Saudi Arabia, yes — many buyers will not order from a new store without it. In the US, UK, Europe and Australia, card payment is normal and COD is rarely needed.
Generally yes: for most online purchases they can cancel within 14 days of delivery without giving a reason. There are exceptions, so check the rules for your products.
A numbers-first way to test products with Meta ads for COD stores: break-even cost per order, test budgets, kill-and-keep rules and how to scale safely.
The shots, hooks, light and sound that make phone-filmed product videos work on TikTok, Reels, Facebook and Snapchat — and the mistakes that make ads expensive.
Revenue is not profit. Here is the full per-order formula, a worked example, break-even ROAS and the costs most new sellers forget.
