How to Prevent Fake COD Orders and Reduce Courier Return Losses in Bangladesh

Cash on Delivery, or COD, is one of the biggest reasons ecommerce works in Bangladesh.

It gives customers a simple way to order products without paying online first. For many shoppers, especially first-time buyers, paying after receiving the parcel feels safer.

But COD creates another problem for ecommerce sellers.

A customer can place an order without paying anything upfront. The seller then spends money to pack the product and send it through a courier. If the customer refuses the parcel, cannot be reached, gives an incorrect address or simply changes their mind, the product comes back.

The seller loses money even though no sale was completed.

For COD-heavy ecommerce businesses, this can become a serious problem. Recent Bangladesh-focused ecommerce sources report COD return rates of around 20%–30% for some stores without proper verification, while other businesses and product categories can experience even higher rates. These figures should be treated as market observations rather than an official nationwide average.

And the cost is not limited to one delivery charge.

There can be:

• Forward delivery cost

• Return delivery cost

• Packaging cost

• Staff processing time

• Lost selling opportunity

• Inventory being unavailable during transit

• Customer-support costs

For planning purposes, many merchants use roughly ৳120–৳150 per failed parcel as a working estimate for courier-related loss, although the actual amount depends on courier rates, delivery zone, parcel size and merchant terms. Some published Bangladesh examples show the total forward-and-return cost can be even higher.

So how can an ecommerce business reduce fake COD orders without making genuine customers uncomfortable?

The answer is not to remove COD.

It is to make COD smarter.

What Is a Fake COD Order?

A fake COD order is an order where the customer provides details and asks the seller to ship the product but does not have a genuine intention to complete the purchase.

Not every failed COD delivery is fraud.

A customer may refuse a parcel because:

• They changed their mind.

• They ordered the wrong product.

• The product arrived later than expected.

• They entered the wrong address.

• They could not receive the parcel.

• They no longer had the money available.

• Someone else in the household placed the order without permission.

There are also deliberate fake orders.

For example, someone may use a real person's phone number and address to place an order as a prank. A competitor or malicious user may repeatedly place orders that are never collected.

The seller usually cannot know the customer's intention when the order arrives.

That is why risk detection is more useful than simply calling every order "fraud."

Why Do So Many COD Parcels Get Returned in Bangladesh?

The problem comes from the way COD changes the customer's commitment.

With prepaid ecommerce, the customer has already transferred money.

With COD, the customer has made a purchase decision without making a financial commitment.

That creates a gap between:

Order placed → Order actually wanted

A customer may feel confident enough to click "Place Order" but change their mind before the courier arrives.

There are several other reasons.

1. Impulse Ordering

Social media advertising makes it very easy to place an order.

A customer sees a product, likes it and places a COD order.

Later, they decide they do not really need it.

2. Multiple Orders

Some shoppers order the same or similar product from several stores.

When the courier arrives, they accept only one parcel.

3. Incorrect Phone Numbers

A customer may accidentally enter the wrong number.

A fake customer may deliberately provide an invalid number.

Either way, the courier cannot reach them.

4. Incorrect Addresses

Bangladesh addresses can be difficult to standardize.

A customer may provide a nearby landmark instead of a complete address, or enter information incorrectly.

The courier may be unable to complete the delivery.

5. Customers Refuse the Parcel

Sometimes the customer simply tells the courier:

"I don't want it anymore."

At this point, the seller has already paid part of the cost of getting the product to the customer.

6. Fake or Malicious Orders

Some orders are placed with no intention of receiving the product.

These are the orders that ecommerce businesses should try to identify before dispatch.

The Real Cost of One Fake COD Order

Consider a simple example.

Suppose a store receives 100 COD orders.

If 25% fail to reach the customer, that means:

25 returned parcels

If the merchant loses an estimated ৳120–৳150 per failed parcel in courier-related costs, the direct delivery loss becomes approximately:

25 × ৳120 = ৳3,000

to

25 × ৳150 = ৳3,750

And this is only the courier-related estimate.

The actual business impact can be higher because the merchant may also lose:

• Packaging material

• Staff time

• Marketing spend

• Product handling cost

• Inventory availability

• Potential profit from a genuine order

Some published Bangladesh examples calculate the total forward-and-return shipping cost at around ৳195 for a failed parcel, demonstrating that actual losses can exceed the lower planning range depending on courier and shipment conditions.

Now imagine the store processes 1,000 COD orders every month.

A 20% return rate means:

200 returned parcels

At ৳120–৳150 each, that represents approximately:

৳24,000–৳30,000

in courier-related losses.

At a 30% return rate:

300 returned parcels

The estimated courier-related loss becomes:

৳36,000–৳45,000

That is money the business could otherwise use for inventory, marketing or growth.

7 Effective Ways to Prevent Fake COD Orders

The best approach is not one security feature.

It is a combination of several small checks.

1. Use AI Fraud Scoring Before Shipping

One of the most useful approaches for a growing ecommerce business is risk scoring.

Instead of treating every COD order equally, a fraud detection system examines different signals and assigns the order a risk level.

For example:

Low risk

A returning customer with:

• Previous successful deliveries

• Verified phone number

• Normal order value

• Consistent address

• Low return history

This order can usually move through the normal fulfilment process.

Medium risk

A new customer with:

• No previous order history

• A higher-than-usual order value

• Unusual delivery information

The business may choose to verify the order before dispatch.

High risk

An order associated with:

• Repeated failed deliveries

• A phone number with a poor order history

• Unusual ordering behavior

• Multiple suspicious orders

• Other high-risk signals

The business can hold the order for review.

Modern ecommerce fraud systems use signals such as order history, customer information, IP or device information, order velocity and other behavioral indicators to score risk. WooCommerce fraud-prevention tools increasingly use this layered approach rather than relying on a single rule.

The important point is:

AI should help prioritize suspicious orders. It should not blindly reject every unusual customer.

A good system gives the merchant another layer of information before shipping.

2. Block Phone Numbers With Repeated Failed Deliveries

Your customer database contains valuable information.

Suppose a phone number has placed:

• 8 COD orders

• 6 failed deliveries

• 5 refused parcels

That number deserves more attention than a customer with a history of successful purchases.

This is why ecommerce businesses should maintain a return-history database.

When an order is returned, the system can record the customer's phone number and delivery history.

If the same number repeatedly creates failed orders, the business can:

• Require verification

• Disable COD

• Ask for an advance delivery payment

• Send the order for manual review

• Block future orders if appropriate

Phone numbers should not automatically be treated as proof of fraud, because numbers can be shared or entered incorrectly.

But repeated behavior is a useful risk signal.

3. Use OTP Phone Verification

One of the simplest ways to improve COD order quality is phone verification.

The process is straightforward.

Step 1

Customer enters their phone number.

Step 2

The store sends a one-time password, or OTP.

Step 3

Customer enters the OTP.

Step 4

The store confirms that the customer has access to that phone number.

This does not prove that the customer will accept the parcel.

But it removes one common problem:

fake or invalid phone numbers.

OTP verification can be especially useful for:

• New customers

• High-value COD orders

• Customers with suspicious order patterns

• Orders from channels with higher fake-order risk

BitCommerz includes an SMS gateway that can be used for order updates and OTP-related customer communication.

4. Add a Small Advance Delivery Charge

You do not always need to ask customers to pay the full product price upfront.

A small advance delivery charge can create a useful middle ground.

For example:

Product price: ৳1,000

Delivery charge: ৳100

The customer pays the delivery charge in advance and pays the remaining product amount when the parcel arrives.

Why can this help?

A person who is not serious about receiving the product may be less willing to make even a small advance payment.

At the same time, genuine COD customers can still use the payment method they prefer.

This approach can be particularly useful for:

• High-return products

• Expensive products

• New customers

• Areas with unusually high return rates

COD fees or partial advance payments are also commonly recommended as a way to reduce the cost of failed COD orders and encourage prepaid payments.

However, the charge should be clearly shown before checkout.

Surprising customers with an unexpected fee can create another conversion problem.

5. Give COD Customers a Reason to Pay Online

The goal should not always be to eliminate COD.

Instead, give customers a reason to choose prepaid payment.

For example:

COD: ৳100 delivery

bKash payment: Free delivery

Or:

COD: Standard delivery

Prepaid: ৳50 discount

The exact offer depends on your margins.

The idea is simple:

Reduce the seller's risk while giving the customer a benefit.

Bangladesh customers may prefer COD because it feels safer. Giving them a small financial benefit can gradually move some genuine customers toward prepaid orders.

6. Verify High-Risk Orders Before Dispatch

You do not necessarily need to call every customer.

That can waste staff time.

Instead, verify orders that have risk signals.

For example:

• First-time customer

• High-value order

• Multiple quantities

• Suspicious phone history

• Previous returned orders

• Unclear address

• Unusually high order frequency

A simple confirmation call can ask:

"Hello, you recently placed a COD order with us for [product]. We are preparing the parcel. Can you please confirm that you still want the order delivered to this address?"

You are not interrogating the customer.

You are confirming purchase intent.

Recent Bangladesh-focused guidance similarly recommends focusing confirmation efforts on orders carrying risk signals rather than manually calling every COD customer.

7. Track Return Rates by Phone, Product and Area

Your return data can tell you where the problem is.

Do not track only:

Total returned orders

Track:

• Return rate by phone number

• Return rate by customer

• Return rate by product

• Return rate by location

• Return rate by courier

• Return rate by marketing source

• Return rate by order value

For example, imagine this data:

Product C deserves investigation.

Maybe:

• The product description is misleading.

• The product quality does not match expectations.

• The price creates impulse orders.

• Customers misunderstand its size.

• The advertising is attracting the wrong audience.

This is important because not every return is a fraud problem.

Sometimes your product page or marketing is creating the problem.

Fake Order Prevention Should Not Hurt Genuine Customers

There is a major mistake businesses can make when fighting fake orders.

They can become too aggressive.

For example, if you block:

• Every new customer

• Every unusual address

• Every high-value order

• Every customer without previous purchase history

you may reduce fake orders.

But you may also reject genuine customers.

A better approach is risk-based verification.

Low-risk order

Ship normally.

Medium-risk order

Request OTP or confirmation.

High-risk order

Ask for partial advance payment or conduct manual verification.

Repeated high-risk behavior

Restrict COD or block the customer according to your business rules.

This layered model is also consistent with modern ecommerce fraud-prevention practices, where suspicious orders are flagged, held or blocked based on multiple signals rather than one simple rule.

Why Manual Fake-Order Checking Stops Working

Manual verification may work when a store receives 10 orders per day.

But imagine receiving:

300 COD orders every day.

Your staff cannot realistically:

• Check every phone number

• Search every customer history

• Call every buyer

• Compare every address

• Check every previous return

• Update every blacklist manually

This is where automation becomes important.

The goal is not to remove people from the process.

It is to let technology identify which orders deserve human attention.

How BitCommerz Helps Reduce COD Fraud and Return Losses

BitCommerz has built COD protection directly into its ecommerce platform.

Its AI Fraud Scoring checks COD orders and assigns a risk score before the order is shipped. BitCommerz says its system uses more than 100,000 data points to evaluate COD orders and can automatically hold high-risk orders before dispatch.

That changes the workflow from:

Order → Ship → Return → Discover the problem

to:

Order → Risk check → Verify or hold → Ship

That difference can save a significant amount of wasted courier expense.

AI Fraud Scoring

Each COD order can be evaluated before shipment.

High-risk orders can be held instead of automatically entering fulfilment.

This gives the merchant an opportunity to verify the customer before spending money on delivery.

COD and Multiple Payment Options

BitCommerz supports COD alongside payment methods such as bKash, Nagad and cards, plus Rocket through SSLCommerz or aamarPay.

This allows a merchant to keep COD available while also giving customers alternatives.

That is important in Bangladesh because removing COD completely can also mean losing genuine customers who prefer to pay after receiving their products.

OTP and SMS Communication

BitCommerz provides an SMS gateway for customer communication, including OTP and order-related messages.

This can support a verification workflow where merchants want additional confirmation before fulfilling selected orders.

Courier Integration

After an order passes the merchant's checks, it still needs to reach the customer.

BitCommerz integrates courier workflows including Pathao, Steadfast, RedX and CarryBee, with parcel booking and tracking from the platform.

This reduces the need to manually copy customer details from one system into another.

Order History and Customer Data

Fraud prevention becomes much more useful when the business can see customer and order history.

BitCommerz provides order management and customer data tools that help merchants keep their ecommerce operations organized.

This creates the foundation for identifying patterns such as repeated failed orders.

A Better COD Workflow for a Bangladesh Ecommerce Store

A practical workflow can look like this:

Step 1: Customer places the order

The customer chooses COD at checkout.

Step 2: The order receives a risk assessment

The system checks available customer and order signals.

Step 3: Low-risk orders continue

The store prepares the parcel normally.

Step 4: Medium-risk orders receive verification

The store can request OTP or customer confirmation.

Step 5: High-risk orders are held

The seller can review the order before shipping.

Step 6: Repeat-return customers receive additional restrictions

The seller can require advance payment or disable COD according to their policy.

Step 7: Approved parcels go to the courier

Courier booking and tracking are handled through the ecommerce workflow.

Step 8: Return data goes back into the system

The business can analyze which customers, products and locations generate the most failed deliveries.

This creates a continuous feedback loop:

Order → Risk → Verification → Delivery → Result → Data → Better Risk Decisions

How Much Can a Store Save by Reducing COD Returns?

Let's use a simple planning example.

Suppose your store processes:

1,000 COD orders per month

and has a:

25% return rate

That means around:

250 parcels return

Using a planning cost of ৳120–৳150 per failed parcel, the estimated courier-related loss is:

৳30,000–৳37,500 per month

Now imagine the store reduces its return rate from 25% to 15%.

Returned parcels fall from:

250 → 150

That means 100 fewer failed deliveries.

At ৳120–৳150 per failed parcel, the potential courier-related saving is approximately:

৳12,000–৳15,000 per month

Again, this is an illustration, not a guaranteed saving. Actual costs vary by courier, zone, parcel type and merchant agreement.

And the total business benefit can be larger because fewer failed deliveries also mean less packaging waste, staff work and inventory disruption.

What Ecommerce Sellers Should Measure

If you want to reduce COD losses, start tracking these numbers every month:

You cannot improve what you do not measure.

Common Mistakes to Avoid

Do Not Disable COD Completely

COD is still important for many Bangladesh customers.

Removing it may reduce fake orders, but it can also reduce genuine sales.

Do Not Trust Every Verified Phone Number

OTP proves that someone controls the phone number.

It does not prove that they will accept the parcel.

Use OTP as one signal, not the entire fraud system.

Do Not Block Customers Based on One Failed Order

A failed delivery does not automatically mean fraud.

The customer may have had a genuine problem.

Look for repeated behavior.

Do Not Call Every Customer

Manual confirmation of every order becomes expensive as the business grows.

Use risk scoring to identify which orders need attention.

Do Not Ignore Genuine Returns

If a product has a high return rate because customers dislike its quality, size or description, fraud detection will not solve the underlying problem.

You need to fix the product or customer experience.

The Bottom Line

COD is not the enemy of ecommerce in Bangladesh.

Unmanaged COD is the problem.

When customers can place an order without paying upfront, the seller carries more delivery risk.

A store with a 20%–30% COD return rate can lose thousands of taka every month simply moving parcels that never become successful sales. The exact rate and cost vary by business, product category, courier and location, so merchants should calculate their own numbers rather than rely on one industry-wide figure.

The solution is to build layers:

AI fraud scoring → phone/OTP verification → customer history → advance delivery charge when appropriate → selective COD restrictions → courier tracking → return analysis

The goal is not to make checkout difficult.

The goal is to make the business smarter about which orders should be shipped immediately and which ones deserve another check.

For Bangladesh ecommerce businesses, that can turn COD from a major source of courier losses into a more controlled and measurable part of the sales process.

BitCommerz brings several of these capabilities into one ecommerce platform, including AI COD fraud scoring, OTP/SMS support, order management, payment options and courier integrations.

The smartest COD strategy is not to stop taking COD orders. It is to stop shipping every COD order blindly.