Cash on Delivery in E-commerce: The Unsolved Problem of Emerging Market Retail

A rider knocks on a door in Lahore. The customer opens it, looks at the package, and says she's changed her mind. The shoes go back in the bag. The rider rides off. The merchant has now paid for fulfillment, packaging, two-way logistics, and the working capital that funded the inventory — for zero revenue.

This happens about 30 to 40 times out of every 100 COD orders in Pakistan, depending on category. Fashion is worse. Electronics slightly better. And every founder running a D2C store in South Asia, MENA, or sub-Saharan Africa knows the number by heart because it's the one that decides whether the business survives the quarter.

COD isn't a payment method. It's a return policy disguised as a payment method. That's the part most outsiders miss.

Why customers still want it (and why that's rational)

When I first started looking at COD e-commerce numbers, I assumed the issue was trust in payment rails. Fix the gateways, expand card penetration, push wallets, and the problem solves itself. I got this wrong.

The real reason a shopper in Karachi or Lagos or Cairo prefers cash on delivery is that it's the only consumer protection mechanism that actually works. Returns are painful. Refunds take weeks. Chargebacks barely exist as a practical tool for the average buyer. So the customer holds the leverage at the door — pay only if the product looks right, fits right, smells right.

Honestly, if I were buying a 4,500 rupee jacket from a brand I'd never heard of, I'd want COD too. It's not irrational behavior. It's a rational response to a market without reliable buyer-side recourse.

This is why the "just educate customers about digital payments" pitch keeps failing. You can't educate someone out of a sensible risk calculation.

The math nobody puts in the deck

Let me walk through what a typical apparel store in Pakistan actually deals with. Average order value sits around 2,800 rupees. Gross margin before fulfillment is roughly 55%. Sounds healthy. Then:

• Forward shipping: ~250 rupees
• Packaging: ~80 rupees
• Payment gateway equivalent for COD (logistics company commission): 2.5% to 4%
• Return-to-origin rate: 32%
• Reverse shipping on RTO: ~200 rupees
• Cash remittance delay from logistics partner: 7 to 14 days

The last point is the killer. You shipped the order. The customer paid. The logistics partner is sitting on your cash for two weeks while you're funding next week's inventory. Working capital becomes the entire game. Brands that grow fast on COD often die fast on COD, because revenue growth and cash position move in opposite directions.

A founder I spoke to in Faisalabad told me his bank balance was lowest on the months his sales were highest. He wasn't joking.

What's actually being tried

A few approaches are genuinely interesting, and a few are theater.

Partial prepayment is the most pragmatic. Ask for 200 to 500 rupees upfront via a wallet, the rest on delivery. RTO rates drop sharply — some merchants report cuts of 40 to 60%. The customer has skin in the game now, even if it's small. The friction at checkout costs you some conversions, but the math usually still wins.

Address verification with AI-driven risk scoring is the second one. Look at the phone number, the address pattern, the order history, the time of day, the device. Flag the high-risk orders for a confirmation call before dispatch. It's not glamorous but it works. Some of the larger logistics players in Pakistan and Egypt are building this in-house now.

Then there's the supply-side problem most D2C brands ignore: their own field operations. If you're a brand selling through both your own site and physical retail (which most growing FMCG and apparel brands eventually are), the disconnect between online RTO behavior and offline retail demand becomes a planning nightmare. Tools like Zivni — which manages field sales teams for FMCG distribution — are starting to matter here because they give brands actual ground-level visibility into what's selling, where, and at what cadence. When your online channel is 35% RTO chaos, knowing that your offline channel moved 1,200 units in Multan this week is the data that keeps inventory decisions sane.

Wallet-based incentives are the third lever. Small discounts (3 to 5%) for prepaid orders. JazzCash and Easypaisa in Pakistan, M-Pesa in Kenya, Fawry in Egypt — the rails exist. The discount eats margin but it's almost always cheaper than the RTO it prevents.

And then the theater: "premium packaging will reduce returns." It won't. The customer who refuses delivery doesn't care about your branded tissue paper. They care about fit, price regret, or the fact that their cousin offered a cheaper alternative yesterday.

The part that won't get solved soon

Here's the thing — COD is downstream of a broader problem. Emerging market e-commerce is being built on top of a logistics and financial infrastructure that wasn't designed for it. Western playbooks assume reliable card payments, fast refunds, and consumer trust in remote sellers. None of those assumptions hold in most of the markets where e-commerce is actually growing fastest.

So we end up with these weird hybrid systems. A WhatsApp catalog. A COD order. A confirmation call. A rider on a motorbike. Cash collected. Cash remitted 12 days later. A return that goes back into stock if the box wasn't opened, or gets written off if it was.

It's inefficient. It's also the only model that currently works at scale for tens of millions of buyers who aren't going to stop shopping just because the system is awkward.

The brands that figure out how to operate inside this mess — not the ones waiting for it to disappear — are the ones building real businesses. The rest are still drawing decks about "digital payment penetration" and wondering why their burn rate keeps climbing.

What does the next version look like? I'm not entirely sure. Probably something closer to escrow than to either pure COD or pure prepaid. Maybe the logistics layer becomes the payment layer. Maybe a few wallets get aggressive enough on buyer protection that customers finally trust the click.

But I'd bet against COD dying this decade. Would you?