India D2C RTO benchmarks 2025: what is normal?

A normal COD RTO (return-to-origin) rate for Indian D2C sits in a steady-state band of roughly 20–35% on cash-on-delivery orders, averaging around 26% in the independent Shipway ShipNotes FY25 dataset, versus under 2% on prepaid. In the festive sale quarter that COD figure climbs sharply, reported as high as ~58% (Unicommerce, 2026). So "is my RTO normal?" really means "is my COD RTO inside that band, and have I measured it in both a quiet month and a festive one?"

What is a normal COD RTO rate in India?

Across the independent Shipway ShipNotes FY25 dataset, COD RTO runs around 26% versus under 2% on prepaid, at ₹180–240 in logistics per returned order. Treat ~26% as the reference point for a COD-heavy brand in a steady month, and the wider ~20–35% band as the range most COD-led brands fall inside once you account for category and geography differences.

Segment	Typical RTO
Prepaid orders	< 2%
COD orders (steady-state)	~20–35% (≈26% avg)
COD orders (festive peak)	up to ~58%

Sources: Shipway ShipNotes FY25 (steady-state COD ~26%, prepaid <2%, ₹180–240/order); Unicommerce India D2C report, 2026 (festive-quarter peak ~58%).

The single most useful thing this table says is that payment method dominates everything else. The jump from under 2% prepaid to ~26% COD is more than a 10x difference, far larger than the spread between any two product categories. That is why a blended RTO number (prepaid and COD mixed together) is almost useless: it hides where the loss actually lives. Always read COD on its own.

A returned COD order costs roughly ₹180–240 in forward-plus-reverse logistics alone (Shipway ShipNotes FY25), before you count the working capital tied up, the unsold inventory cycling back, and the margin you never booked.

How much worse does RTO get in the festive quarter?

RTO does not stay flat across the year. During the festive sale window, Big Billion Days, Amazon's Great Indian Festival, the run-up to Diwali, COD RTO has been reported as high as ~58% (Unicommerce, 2026). That is more than double a normal-month figure, and it arrives in the exact quarter when order volume is highest, so the rupee impact compounds twice over: more orders, each carrying a higher bounce probability.

The drivers are structural, not random. Discount-led demand pulls in lower-intent and first-time buyers who are more likely to over-order, change their mind, or refuse at the door. Impulse purchases on heavily marked-down items have weak commitment behind them. Address and serviceability problems that exist all year simply multiply when volume spikes. The practical lesson: if you only ever measure RTO in a calm month, you will badly under-estimate your festive exposure, which is precisely when the cost is largest and the cash-flow hit lands hardest.

What drives the differences between brands?

Two brands selling similar products can sit at opposite ends of the ~20–35% band. The published benchmarks are averages; your own number is moved by a handful of qualitative factors. We deliberately avoid putting invented percentages on these, the honest position is to describe the direction each lever pushes, not to fabricate a precise figure the data does not support.

• COD share. The higher your COD mix, the higher your blended RTO, simply because COD carries roughly 10x the bounce rate of prepaid. Shifting even the riskiest orders to prepaid moves the average meaningfully.
• Category. Fashion, footwear and accessories tend to sit toward the upper end of the band, sizing uncertainty and impulse buying both raise refusals. Considered, higher-ticket or replenishment purchases tend to sit lower. We describe this as a direction, not a number; published category-level percentages vary too much to quote responsibly.
• Price point and AOV. Very low-AOV impulse orders refuse more often, because the buyer has little at stake in declining at the door. Higher-consideration purchases generally show more commitment.
• Geography and tier. Address quality, last-mile serviceability and delivery reliability vary widely across tier-2 and tier-3 pincodes, and that variance feeds straight into RTO.
• Address and data quality. Incomplete addresses, unreachable phone numbers and duplicate orders bounce more. Cleaning this at checkout is one of the cheapest levers available.
• Whether you intervene before dispatch. Brands that confirm and risk-score COD orders sit well below the benchmark; brands that ship everything blind sit at or above it. This is the lever you control most directly.

How do you calculate your own RTO cost?

The benchmark only matters once you translate it into rupees on your order book. Here is the illustrative formula, the same logic the RTO calculator runs:

Monthly RTO cost ≈ Orders × %COD × %COD-RTO × ₹ per bounce

1. Start with your monthly order count. Use a real number from your own dashboard, not an estimate.
2. Multiply by your COD share (%COD) to isolate the orders actually exposed to COD RTO.
3. Multiply by your COD RTO rate (%COD-RTO). If you do not yet know your own, ~26% is a defensible placeholder for a normal month; use a higher figure for a festive month.
4. Multiply by the cost per bounce, ₹180–240 from Shipway ShipNotes FY25, to convert bounces into rupees.

To make the mechanics concrete, and this is an illustrative, modeled example, not a benchmark, take a brand doing 10,000 orders a month at 60% COD. That is 6,000 COD orders. Apply a modeled 26% COD RTO and you get ~1,560 bounces; at ₹200 each that is roughly ₹3.1–3.7 lakh a month in pure logistics drag, before working capital and lost margin. Swap in a festive-quarter rate and the same brand's exposure can more than double. Run your own figures through the calculator rather than trusting the illustration, the numbers above exist only to show the shape of the math.

1. Measure COD RTO separately from prepaid, a blended number hides the problem.
2. Pull at least 60–90 days so you are not fooled by one good or bad week.
3. Look at a festive month and a normal month separately, because the gap between them is where most brands are caught out.

Above the benchmark? That is the opportunity

If your COD RTO is sitting north of ~26% in a normal month, that gap is recoverable margin, not a fact of life. The interventions that close it, confirmation, risk-scoring, prepaid nudges, NDR recovery, are covered in what COD RTO is and how WhatsApp confirmation works. Whether you run them yourself or hire them out is covered in managed vs DIY.

One caveat worth repeating: these are external benchmarks and modeled figures. The only number that truly describes your brand is the one measured on your own orders, which is what HootMonk's 2-week pilot exists to produce. If you want that number read against the benchmark on your real data, get in touch.