D2C ProfitCalculate My Profit

How to reduce RTO on COD orders: 8 fixes that work

Practical ways Indian sellers cut COD RTO, from order confirmation and prepaid incentives to faster delivery, and what a lower RTO rate is worth in rupees.

Published 26 Oct 2026

Key takeaways

  • Most RTO comes from COD, so most fixes are about COD: confirm it, discourage it, or restrict it.
  • On a 500-order store, cutting COD RTO from 26% to 18% is worth ₹14,877 a month.
  • Moving buyers to prepaid pays even after a 5% discount — up to a break-even of 19.4% off on this product.
  • Faster delivery and same-day action on failed delivery attempts save orders that would otherwise come back.

Most RTO in India comes from Cash on Delivery: about 26% of COD orders come back undelivered, against under 2% of prepaid ones, and fashion runs closer to 45% on COD. You can't make COD buyers behave like prepaid ones, but you can stop paying to ship orders that were never going to be accepted. Here are the fixes that work, and what they're worth in rupees.

First, what a lower RTO rate is worth

Worked example

A ₹999 apparel store doing 500 orders a month, 60% of them COD, starting at a 26% COD RTO rate. Bringing COD RTO down to 18% — and changing nothing else:

Inputs

Monthly orders
500
COD share
60%
COD RTO, before → after
26% → 18%

Result

Contribution profit before
₹1,41,242
Contribution profit after
₹1,56,119
Extra profit every month
₹14,877

That's the prize. Here's how to go after it.

Measure before you fix: where is your RTO coming from?

Before changing anything, break last month's RTO down four ways. RTO is rarely spread evenly, and where it clusters tells you which fix to start with:

  • By payment mode — COD vs prepaid RTO rate.
  • By pin code or city — check whether a few areas stand out.
  • By product — sizing problems or misleading photos show up as product-specific RTO.
  • By NDR reason — "refused", "not reachable" and "address issue" each point to a different fix.

1. Confirm COD orders before you ship them

Send every COD order a WhatsApp message, call or OTP asking the buyer to confirm, and don't dispatch until they do. An order cancelled before shipping costs you almost nothing; the same order refused at the door costs you freight both ways. This turns expensive RTO into cheap cancellations, and it's one of the cheapest fixes to set up.

2. Make prepaid the easier choice

Offer a small discount or cashback for paying online. Every buyer you move from COD to prepaid is worth real money — on the store above, about ₹124 per order even after a 5% discount, so converting 100 COD orders a month adds ₹12,427. Keep the discount below the break-even point — 19.4% here — and check your own with the Prepaid vs COD Calculator. Partial COD (a small amount paid upfront, the rest on delivery) works on the same idea: any commitment at all makes a buyer far less likely to refuse the parcel.

3. Ship and deliver faster

The longer a buyer waits, the more likely they are to change their mind. GoKwik's data shows orders delivered in one to two days seeing about 22% RTO, against roughly 35% when delivery slips past five days. Dispatch the same or next day, and pick couriers by their actual delivery speed in your top pin codes, not just their rate card.

4. Fix addresses and phone numbers at checkout

Check pin-code serviceability before the order is placed, make landmark and phone fields mandatory, and validate phone numbers. A surprising share of RTO is simply couriers who can't find the address or reach the buyer.

5. Restrict COD where it keeps failing

Look at your RTO by pin code and by customer. Where COD keeps coming back — specific pin codes, repeat refusers, unusually large COD orders — offer prepaid only, or ask for a partial payment upfront.

6. Act on failed delivery attempts the same day

When the courier reports a failed attempt (an NDR — non-delivery report), call or message the buyer immediately and schedule a re-attempt. Many orders marked "customer not available" can still be saved if someone follows up before the courier gives up and returns the parcel.

7. Set honest expectations on the product page

Accurate photos, size charts and a realistic delivery date mean fewer buyers who see the parcel and decide it isn't what they wanted. This cuts customer returns as much as RTO.

8. Consider a small COD fee

A modest charge for paying on delivery nudges some buyers to prepaid and filters out the least serious ones. Watch your conversion rate when you try it — if COD orders drop more than RTO savings make up for, roll it back.

Where to start

  • Measure your COD and prepaid RTO rates separately and compare them with the benchmarks for your category.
  • Put a monthly rupee figure on your RTO with the RTO Loss Calculator.
  • Start with COD confirmation and a prepaid incentive — both are cheap to set up and easy to measure. Then check COD still makes money on its own with the COD Profit Calculator.
FAQ
How can I reduce RTO for COD orders?

Confirm every COD order by WhatsApp, call or OTP before dispatch, offer a discount for paying online, ship fast, validate addresses and phone numbers at checkout, restrict COD for high-risk pin codes, and act on failed delivery attempts the same day.

What is COD order verification?

Asking the buyer to confirm a Cash on Delivery order — usually by WhatsApp message, automated call or OTP — before you ship it. Unconfirmed orders are cancelled before dispatch, which costs almost nothing compared with an RTO.

What is NDR management?

Following up on non-delivery reports — the courier's record of a failed delivery attempt. Contacting the buyer quickly and scheduling a re-attempt can save an order before it turns into an RTO.

How much discount should I give for prepaid orders?

Less than the break-even discount at which a prepaid order earns the same as a COD one. Use the Prepaid vs COD Calculator to find yours.

Should I stop offering COD to reduce RTO?

Usually not outright — many buyers only order with COD. It's safer to keep COD, verify orders, reward prepaid, and limit COD only where it keeps failing.