COD vs Prepaid Orders

Prepaid means the customer pays (or authorizes payment) before you commit inventory and a label. Cash on delivery (COD) means the courier or your staff collects cash (or a local cash-equivalent) when the parcel is handed over.

Neither is morally better. They optimize different risks. In some regions COD is a large share of e-commerce; in others it is rare. This article is operational, not a campaign for or against COD.

Shipping physics are the same—How E-Commerce Shipping Works. Payment stack for prepaid is E-Commerce Payment Gateways Explained. Checkout declines are Payment Failures and Checkout Recovery.

Conversion vs RTO and rejection

COD often raises checkout conversion because the customer does not complete a card or UPI flow. The leak moves downstream:

• Customer refuses the parcel.
• Address is fake or unreachable.
• Partial rejection (opened, then refused).
• Cash not ready; second attempt fees.

RTO (return to origin) means you paid outbound (and often inbound) for zero revenue, plus refurbishment if the pack was opened.

Prepaid leaks at authorization: OTP, insufficient funds, 3-D Secure, issuer declines. Those are painful but you usually have not shipped yet.

Use the conversion rate calculator on placed orders and a second funnel on delivered and paid. If you only watch CVR, COD will always look like a win.

Cash flow

	Prepaid	COD
Cash timing	At order (minus settlement delay)	After delivery, minus courier remittance lag
Working capital	Better for inventory	You fund COGS and freight first
Fraud / fake orders	Card/wallet fraud, friendly fraud	Fake addresses, habitual refusers
Reconciliation	Gateway settlement	Courier COD payouts, short cash, disputes

High-ticket COD is a working-capital product. Low-ticket COD can still lose if RTO is high: the shipping cost calculator on a round trip is the real “discount.”

When COD is rational

COD can be a fit when:

• Your market’s shoppers will not prepaid a new brand at your ticket size.
• Category return/refusal rates are manageable (replenishment, known SKUs) rather than fashion-try-on at the door.
• You can score addresses and customers (limit COD for new customers, high-RTO postcodes, or after N refusals)—within the law and your platform’s rules.
• Courier remittance is predictable enough to run payroll.

COD is a poor default when:

• Contribution after a typical RTO rate is negative—run the margin math.
• You cannot staff verification calls or quality addresses.
• The SKU is easily resold or high-theft.

Prepaid-only is rational when trust is already there (repeat buyers, brand, low ticket with wallets) or when RTO would dominate.

Operations that keep COD from eating the P&L

• Confirm orders (call/SMS) on first-time COD if it reduces fake addresses enough to pay for itself.
• Cap COD order value.
• Split methods: prepaid incentives (small, contribution-safe) vs COD as default.
• Fast dispatch: delay increases refusal.
• Honest ETAs: late COD is refused COD.

Do not harass customers or collect extra identity documents “to reduce RTO” beyond what you need to fulfill and what the law allows. That is a trust and compliance problem, not a growth hack.

Measurement

Dashboard (minimum):

• Share of orders COD vs prepaid.
• RTO / refusal rate by method, postcode, SKU.
• Contribution per delivered order by method.
• Time to remittance.

AOV is often higher on COD and then collapses after refusals—report delivered AOV.

Prepaid incentives without wrecking contribution

A small prepaid discount (or free shipping on prepaid only) can shift mix if the discount is less than expected RTO cost on COD. Run the numbers: if RTO costs $9 on a typical COD order and a 3% prepaid discount is $1.80 on a $60 ticket, the incentive can be rational. A 20% prepaid coupon that destroys margin is not.

Never surprise a COD customer with extra ID checks at the door that were not in the checkout. That is a refusal generator.

Address quality matters more than a pep talk: incomplete buildings, fake numbers, and “office until 5pm” on an evening-only courier. If you cannot verify, prepaid or a smaller COD cap is safer than hoping the first mile works.

Courier contracts differ on who pays RTO freight and how cash is remitted. Read the settlement schedule before you treat COD GMV as this week’s cash.

First-time vs repeat

Repeat buyers often prepaid more readily. Forcing COD-only because “it converts” trains a cash-cycle you may not want as the brand matures. Offer both, default by customer type if your platform allows it lawfully, and watch delivered contribution—not checkout CVR alone.

Fake-address rings and serial refusers are an operations problem: cap COD, require prepaid after N refusals (where lawful), and do not ship high-theft SKUs COD. That is risk control, not harassment.

Partial delivery (opened, then refused) should be a photographed exception with the courier, not an automatic full refund plus free restock of a used unit—within your published policy and local consumer rules. Train the warehouse on “refused, unopened” vs “refused, used” so finance and CX use the same codes. Do not treat courier short-cash as a marketing variance; it is a settlement control.

The shipping guide and conversion guide cover the adjacent playbooks. Choose COD when the alternative is no order and delivered contribution still works. Choose prepaid when the door is where you currently lose the business.