How to Reduce RTO Without Removing COD: 5 Fixes and Costs
You can cut RTO and keep cash on delivery. Here are five fixes beyond prepaid offers, what each one costs, where each one stops working, and a test to check it on your own orders.
By Nitish KumarOct 29, 2026 • 6 min read
The usual advice for high RTO is to switch off cash on delivery. For many Indian stores that also switches off a large part of the market. This post is for the store that wants to keep COD. It covers five fixes that are not in RTO in E-commerce: COD vs Prepaid, and for each one it says what the fix costs and where it stops working. I could not find a trustworthy published effect size for any of them, so I do not quote one. Instead I give you a way to test each fix on your own orders.
How do you know a fix is worth it?
Every fix has two sides. It prevents some RTOs, and it may also lose some genuine orders. Use this test:
Net gain = (RTOs prevented × C) − (genuine orders lost × M) − the cost of running the fix
- C: the full cost of one RTO: forward shipping, return shipping, and packaging and handling.
- M: what you earn on a delivered order after product cost, shipping and fees, before ad spend.
I explain how to work out both in What Is a Good RTO Rate?. Example, not real data: with C = ₹130 and M = ₹400, a fix has to prevent about three RTOs (400 ÷ 130 = 3.1) for every genuine order it loses, before it covers its own running cost. Keep that ratio in mind as you read the five fixes.
1. Fix the address before it ships
What it is. Make house number, street, landmark and PIN code required at checkout, and check the PIN code against a valid list. Hold orders with an obviously weak address until someone corrects them.
What it costs. A little checkout friction, and a few minutes of someone's time for each held order. Your checkout settings or a checkout app decide how much of this you can automate, so check what yours supports.
Where it stops. It catches missing and malformed addresses. It does not catch an address that looks complete but is wrong, or a buyer who simply will not be home. A Shipway report names address accuracy as one factor behind city-level differences in RTO, but gives no number for its effect.
2. Work the NDR queue within a day
What it is. An NDR (non-delivery report) is the courier's note that a delivery attempt failed. Look at the queue every day. For each failed attempt, reach the customer, confirm the address and a time, and ask the courier to re-attempt.
What it costs. Staff time every day, plus any extra attempt fee your courier charges. Check your courier contract for both.
Where it stops. It only saves orders where the customer still wants the parcel. It cannot fix a changed mind, a courier that never attempted delivery, or a PIN code the courier does not serve. Your courier also decides how many attempts it makes before it sends the order back, so read its policy.
3. Speed up the first attempt
What it is. Dispatch the day the order is placed, and send each order with the courier that is fastest on that route.
What it costs. Faster services can cost more per parcel, and same-day dispatch needs stock and packing ready.
Where it stops. The evidence is a correlation. In the Shipway FY25 data, reported by Media Infoline, COD RTO was 22% when the delivery attempt came within 1 to 2 days, 27% at 3 to 5 days and 35% after more than 5 days. The report reads this as slower delivery causing more RTO, but it states no method, so slower deliveries may also be to harder places. Test it on your own orders before you pay more for speed.
4. Ask for a small advance on risky COD orders
What it is. For orders you rate as risky, such as a first order or one with a weak address, ask for a small part-payment up front and collect the rest in cash.
What it costs. Some buyers will drop out, and each one is a lost M. You also pay gateway fees, and you have to refund the advance when an order is cancelled. Whether your store can take a part-payment depends on your checkout and payment gateway, so check before you plan around it.
Where it stops. An advance filters out casual orders, but it does not stop a buyer who paid and still refuses. Apply it to a defined group of orders, not to everyone, or it turns into the COD removal you wanted to avoid.
5. Cap COD by order value or buyer history
What it is. Allow COD only below a set order value, or only after a buyer has taken a prior order.
What it costs. Every blocked buyer who would have taken delivery costs you M. In smaller cities, where COD is often the only comfortable way to pay, that loss can be large.
Where it stops. Order value is a weak signal. In the same Shipway data, COD RTO by order value was 25% under ₹500, 28% from ₹500 to ₹1,000 and 24% above ₹1,000. The spread is small, so a value cap would cut good and bad orders alike. Buyer history is a stronger signal because it is your own data. Start there.
What should you try first?
Start with the cheapest fix that does not lose orders: the address check and the daily NDR queue. Measure your RTO rate for a month, split by payment type. Add a fix that can lose genuine orders (an advance or a cap) only for a defined group, and compare that group with the rest. To see what each point of RTO is worth in rupees, use the free RTO Cost Calculator. If you are not sure where your RTOs come from, read 5 Common Reasons for RTO in E-commerce first.
How to reduce RTO without removing COD: frequently asked questions
Can I reduce RTO without removing COD?
Yes. Fixing the address, working failed deliveries daily and speeding up the first attempt do not take COD away from anyone. Fixes that add friction, such as an advance or a cap, cost you some genuine orders, so test them on a small group first.
Which fix works best?
I found no trustworthy published figure that ranks them. The best fix for you depends on where your RTOs come from, so find the cause first, then test one change at a time.
How do I know if a fix lost me more than it saved?
Use the net gain test above. Count the RTOs it prevented and the genuine orders it lost, and weigh them with C and M. If you cannot tell the two groups apart, compare a tested group with a control group over the same weeks.
Your action checklist
- Work out C and M for your store.
- Make the address fields and the PIN code check required at checkout.
- Set a daily time to clear the NDR queue.
- Compare your RTO rate by delivery speed.
- Pick one fix, apply it to one group of orders, and compare it with the rest after a month.
Your next step: Work out C and M for your store today, so every fix you test has a number to beat.
Sources: Shipway ShipNotes FY25, as reported by Media Infoline (29 Jul 2025), for the delivery-speed, order-value and address-accuracy points.