What actually causes RTO in Indian ecommerce?
RTO happens for three different reasons that most brands treat as one: unreachable or fake orders, structural delivery failure in specific pincodes and couriers, and genuine customers who changed their mind at the door. Verification tools solve the first. Courier data solves the second. The third is a trust problem, and it is largely driven by whether that customer has had a good experience with you before, which is what DOPE reads for Shopify and D2C brands.
The scale is brutal. India's D2C RTO rate averages 20 to 30%, against a global benchmark of 8 to 12%, and COD-heavy categories like fashion and footwear touch 40% (2026 industry data). Every RTO costs ₹150 to ₹300 before you count the ad spend that acquired that customer, and industry estimates put collective annual losses to Indian D2C above ₹8,000 crore. Here is why the standard playbook only takes you so far.
The COD gap is the whole story
Start with the number that reframes everything. COD orders return at roughly 26% nationally. Prepaid orders return at under 2%. COD is about 13x worse (2026 industry data).
And the concentration is worse than the average suggests: COD represents around 62% of order volume but drives 76 to 83% of total RTO volume. Meanwhile COD is not going anywhere, it is 60 to 65% of Indian ecommerce, and higher in Tier 2 and Tier 3 cities, which is exactly where 66% of new D2C orders are now coming from.
So the problem is structural. You cannot switch off COD without losing a large share of conversions, and you cannot ignore the tier 2 and 3 growth where COD dominates. You have to make COD orders stick.
Why verification alone hits a ceiling
The standard RTO playbook is correct as far as it goes: validate addresses at checkout, confirm COD orders before dispatch, manage NDRs fast, nudge customers to prepaid, route around weak couriers. Do all of it. Brands running this discipline consistently hold RTO under 12%.
But notice what verification actually catches. It catches the fake order, the wrong address, the unreachable phone number. It filters out the orders that were never real. What it does not catch is the customer who was completely real, gave a correct address, answered the confirmation, and then refused the parcel at the door anyway.
And the verification layer itself is weakening. In 2026, confirmation phone calls increasingly fail simply because people do not answer unknown numbers. The industry's default RTO tactic is running into a behavioural wall, which means the brands that win will be the ones who reduce the need for verification rather than doing more of it.
The refusal at the door is a trust decision
Here is the part the RTO conversation misses. A customer who accepts a COD parcel is paying money to a stranger at their doorstep for something they have not inspected. That is a trust transaction, not a logistics one.
The customer who refuses did not usually decide at the door. They decided somewhere between clicking buy and the knock, and the biggest input into that decision is what they already believe about your brand. A first-time buyer with no relationship has no reason to feel committed. A repeat customer whose last order arrived damaged has a very good reason to refuse this one. Only about 1 in 26 unhappy customers ever complains, so most of those customers never told you they were disappointed. They just declined the next delivery instead.
That is why RTO and churn are the same problem wearing different clothes. A silently dissatisfied customer shows up as a lapsed reorder if they were prepaid, and as an RTO if they were COD. Same broken trust, two different line items on your P&L.
The pincode and courier layer nobody reads properly
The other half of RTO has nothing to do with the customer at all.
Pull your last month of returns and sort by delivery area. Almost every brand finds three to five pincodes responsible for a disproportionate share of RTOs. The same holds for couriers: a partner performing fine nationally can be quietly failing in specific zones, and rerouting those zones to a different carrier measurably improves delivery success.
This is a structural failure being misread as a customer failure. Those customers were often willing to accept the parcel. The delivery attempt was simply mishandled, made once at a bad hour, or never made at all. Blaming intent when the cause is courier performance means you keep paying for the same failure every month.
How DOPE reads the RTO signals your logistics stack can't
DOPE is a customer intelligence tool for Shopify and D2C brands, and it works on the two RTO layers that verification and shipping tools do not touch.
The first is courier and pincode risk. DOPE surfaces where your deliveries are structurally failing, which pincodes and which courier partners are quietly draining your margin, so you can route around the problem instead of absorbing it as a national average.
The second is the customer trust layer. DOPE reads behavior and sentiment across your customer base, so you know which customers have had a poor experience with you before. That is the single best predictor of whether the next COD parcel gets accepted or refused. A customer whose last order disappointed them is a refusal waiting to happen, and right now most brands have no way of knowing that before they ship.
Put together, that changes the question from "was this order verified" to "does this customer have a reason to accept it." Verification confirms the order exists. Trust decides whether it gets taken.
A note on how DOPE works: it surfaces the risk patterns and the at-risk customers, then you act using your own stack and your own channels, your WhatsApp, email, or SMS, in your own voice. DOPE does not call or message your customers, and it does not replace your verification or shipping tools. It is the intelligence layer that tells you where the failures are concentrated and which customers no longer trust you enough to pay at the door.
The brands holding RTO under 12% are not just verifying harder. They are the ones whose customers actually want the parcel. For the wider view on silent dissatisfaction, see the customers who leave without a word, and for what returns are really telling you, most of your returns aren't fraud.
FAQ
What is a good RTO rate for Indian D2C brands?
The national average sits at 20 to 30%, with COD-heavy categories like fashion and footwear reaching 40%. Anything under 12% is considered best-in-class for Indian D2C. Benchmark against your category and your own trend rather than the national figure.
Why is COD RTO so much higher than prepaid?
COD returns at around 26% versus under 2% for prepaid, roughly 13x worse, because COD requires no commitment at purchase. The customer can refuse at the door with nothing lost. Accepting a COD parcel is a trust decision, so prior experience with your brand heavily influences it.
How do I reduce RTO in ecommerce?
Layer three things. Verify addresses and COD orders before dispatch, route around underperforming pincodes and couriers, and reduce the trust gap that makes customers refuse in the first place. DOPE surfaces the courier and pincode risk plus the customers whose prior experience makes refusal likely.
Do COD confirmation calls still work in 2026?
Less than they used to. Customers increasingly do not answer unknown numbers, so call-based verification is losing effectiveness. This is why reducing the underlying reason for refusal matters more than adding another verification step on top.
Can DOPE stop RTO orders before they ship?
DOPE does not verify orders or contact customers. It surfaces where RTO is concentrated by courier and pincode, and which customers carry a prior bad experience that makes refusal likely, so your team can act through your existing verification stack and your own channels.
Top comments (0)