How to Reduce RTO: Twelve Tactics, Ordered by What They Return

Every article on RTO lists thirty things you could do. This one lists twelve you should, in the order they pay back, with the number each one typically moves. Start at the top; most sellers halve their RTO before reaching number six.

1. Confirm COD orders before you ship

The single largest lever. A WhatsApp message with the order summary and a one-tap confirm, or an IVR call, within an hour of the order. Orders that do not confirm within 24 hours are held and re-messaged once; still unconfirmed, cancelled.

Typical effect: COD RTO from 22 percent to 13 to 15 percent. Cost: a rupee or two per order in messaging. Some orders are lost at the confirmation step; almost all of them were RTOs waiting to happen.

2. Work every NDR the same day

A failed delivery attempt is not an RTO yet. The courier will try again in a day or two. If the buyer has been reached in between, with the address corrected or a delivery time agreed, the second attempt succeeds. If not, it fails the same way.

Set up NDR notifications, and have someone (or an automation) contact the buyer within hours. On Courier Uncle, NDR automation re-attempts the first attempts automatically and pushes the instruction to the courier; a person handles only the escalations.

Typical effect: 40 to 60 percent of NDRs converted to deliveries, which is a third off the RTO rate on its own.

3. Validate the address at checkout

Pincode must exist and be serviceable. House or flat number mandatory. Phone number verified by OTP. Addresses under 20 characters rejected with a prompt. Cheap, one-time, and it removes the “could not locate” class of RTO entirely.

Typical effect: 2 to 4 points off RTO.

4. Nudge to prepaid

A visible discount for paying now: 3 to 5 percent, or free shipping on prepaid only, or a small gift. Every order that converts is an order that cannot be refused at the door.

Typical effect: prepaid share from 35 percent to 50 to 55 percent within a quarter, which at a 3x RTO differential is 3 to 5 points off blended RTO. The discount costs less than the RTOs it prevents; the arithmetic is here.

5. Ship faster on long lanes

Each day in transit is a day for the buyer to change their mind. On Zone D and E, prefer the courier with the shortest transit even if it costs Rs 10 more; on Zone A and B it rarely matters. Courier rules on an aggregator can do this per zone automatically.

Typical effect: 1 to 3 points on long lanes.

6. Partial COD on higher values

Collect Rs 99 to 199 online at checkout on COD orders above a threshold; the rest on delivery. The buyer with money already in the order does not refuse it.

Typical effect: COD RTO on orders above Rs 1,500 from 25 percent to under 10. Some checkout drop-off; net positive for most sellers above the threshold. Implementation notes here.

7. Block repeat refusers

A buyer who has refused two COD parcels gets prepaid-only at the next checkout. A phone number and address hash is enough to recognise them. Small share of buyers, outsized share of RTOs.

Typical effect: 1 to 2 points.

8. Cap COD by value and pincode

Turn COD off above the order value where the fee plus expected RTO cost exceeds your margin, and on pincodes where your own history shows RTO over 30 percent. The RTO report by pincode tells you which.

Typical effect: 1 to 3 points, concentrated in the worst lanes.

9. Send tracking proactively

A message at pickup, in transit, out for delivery. A buyer who knows the parcel arrives tomorrow is home tomorrow. Also cuts “where is my order” tickets.

Typical effect: 1 point, plus support load.

10. Get the rider’s call answered

Delivery attempts fail when the buyer does not pick up an unknown number. Tell the buyer in the out-for-delivery message that the courier will call, and from roughly what number range. Some couriers offer a masked number the buyer has seen before.

Typical effect: under 1 point, but nearly free.

11. Fix the product page

RTO caused by “not what I expected” is a listing problem. Accurate photos, sizes, and a size chart cut both RTO and post-delivery returns.

Typical effect: varies by category; large for apparel and footwear.

12. Reconcile and repeat

Monthly: RTO by courier, zone, pincode, product and payment mode. The cut that is worst this month is the tactic to push next month. The report worth building is here.

What the stack looks like together

A seller starting at 24 percent blended RTO who does 1 through 6 typically lands at 9 to 11 percent inside two quarters. The cash effect on 3,000 orders a month at Rs 250 per RTO is roughly Rs 1 lakh a month, before counting the margin on the orders that now deliver.

Frequently asked questions

Which tactic should I do first?

Order confirmation. It has the largest effect, the lowest cost and can be running this week.

Will confirmation calls annoy buyers?

A single WhatsApp with the order summary and a confirm button does not. Repeated calls do. One message, one reminder, then hold.

Does NDR automation replace a person?

For the first one or two attempts, yes: the re-attempt instruction goes to the courier automatically. Escalations (a wrong address, a buyer who wants to cancel) still need a human, but there are far fewer of them.

How much prepaid discount is right?

Between the COD fee and the expected RTO cost per COD order, which for most sellers is Rs 40 to 80 on a Rs 1,000 order: 3 to 5 percent. More than that gives away margin; less does not move buyers.

Should I stop COD entirely?

Almost never. COD is a large share of Indian demand. The goal is to offer it where it earns its cost and to make prepaid attractive everywhere else.

How do I measure whether a tactic worked?

Compare RTO on the affected segment before and after, over at least four weeks, holding the others constant. The RTO report by cut is built for this.

Ship smarter with Courier Uncle

NDR automation, RTO by every cut, and courier rules that pick the fastest carrier on long lanes. Start free.