COD vs bKash Advance: How Bangladeshi Sellers Reduce Fake Orders

Every online seller in Bangladesh knows the feeling: the courier calls, the customer's phone is off, and your parcel is coming back. You pay the delivery charge both ways, the product sat in transit for a week, and you earned nothing. Fake and refused COD orders quietly eat 10–30% of many sellers' profit, and most sellers treat it as fate. It isn't. It's a system problem with system fixes.

Why you can't just drop COD

Cash on delivery exists because it solves the customer's trust problem: they've been burned by pages that took bKash advance and vanished, or delivered something very different from the photo. COD says "see the product, then pay." Drop it entirely and your conversion falls off a cliff, especially with new customers. The goal is not to eliminate COD; it's to filter out the orders that were never real.

Why fake orders happen

• Impulse without commitment, ordering costs the customer nothing, so people order casually and change their minds.
• Wrong numbers and addresses, typos, or someone ordering "for a friend."
• Pranks and rivals, rare but real.
• Price shopping, they ordered from three pages and accept whichever arrives first.

The filter ladder: five fixes, cheapest first

1. Confirm every order by phone or WhatsApp

Before booking the courier, send one message: "Confirming your order: [product], total ৳X, delivering [day]. Reply YES to confirm." No reply within a few hours → one phone call → still nothing → don't ship. This one habit removes a large share of fake orders and costs you nothing.

2. Advance delivery charge outside Dhaka

The single most effective filter used by experienced sellers: COD inside Dhaka, but the delivery charge (৳100–150) paid in advance via bKash for outside-Dhaka orders. The logic: a customer unwilling to commit ৳130 was never going to hand ৳1,500 to the courier. You're not asking for the product price, just proof of intent. Genuine customers rarely object when you explain it politely.

3. Full or partial advance for made-to-order and high-value items

Custom items (tailored clothes, engraved gifts, pre-orders) and anything expensive should take 20–50% advance. A returned custom product is a total loss, you can't resell it easily.

4. Track repeat offenders

Keep a simple list of phone numbers that refused delivery. Before shipping a new order, check it. Couriers like Steadfast and RedX also expose customer success rates, a number with three previous refused parcels across the network is a red flag.

5. Make ordering slightly more deliberate

Orders that arrive as a vague comment ("inbox koren") flake far more than orders where the customer walked through a real checkout, chose the product, entered their address, saw the delivery charge and total. A structured order is a small commitment device. This is one quiet advantage of having a store link instead of taking orders purely from comments: by the time the order reaches your WhatsApp, the customer has already confirmed the details themselves, and you have their address in writing.

A sane default policy

Measure it

Track one number monthly: return rate = parcels refused or undeliverable ÷ parcels shipped. Above 15% means your filters are too loose. Below 5% with falling sales means they may be too tight, loosen for repeat customers first. Most sellers who apply the ladder above land at 5–10%, which turns directly into profit: at 100 parcels a month with ৳120 two-way courier cost, cutting returns from 25% to 8% saves roughly ৳2,000 in courier fees alone, before counting the recovered sales.