E-Commerce Returns and RTO in Indian Fashion 2026: Data, Costs and What Reduces Them
Every fourth or fifth parcel shipped by an Indian fashion e-commerce brand comes back. Some of it is the classic return — tried, disliked, sent back. Much of it never reaches the customer at all: refused at the door, unreachable, cancelled in transit — the uniquely COD-shaped failure Indian logistics calls RTO, return to origin. This report compiles the verified 2023-2026 data on both: how big the problem is, what each bounce costs, why size and fit dominates the causes, and which interventions measurably shrink it.
Key findings
- Fashion and apparel returns in India run at 25-40% per GoKwik — the highest of any e-commerce category — while Pragma pegs fashion and footwear at 30-35% against a ~17% all-category average.[6]
- RTO fell from 39.2% at the November 2025 festive peak to 21.0% by March 2026 among optimised D2C brands, on a dataset of 410 million+ shipments (Unicommerce).[1]
- During the festive quarter, 58% of COD orders came back — prepaid orders returned at under 15% (Unicommerce India D2C Report 2026).[1]
- Across FY23, COD orders returned at 20.9% vs 5.8% for prepaid — a 3.6x gap that makes payment mix the single biggest returns lever (Unicommerce).[3]
- Every RTO parcel costs an Indian brand Rs 150-300 in direct logistics and Rs 450-900 fully loaded once ad spend and handling are counted (ClickPost).[9]
- Size and fit is the #1 driver: 53% of apparel returns globally stem from fit (Coresight / PRIME AI), and 67-73% of Indian shoppers pick sizes on guesswork rather than measurements (Pragma).[11]
- Returns are an environmental cost too: e-commerce returns generate up to 24 million metric tons of CO2 a year and sent 9.5 billion pounds of product to landfill in 2022 (Optoro).[18]
What's in this report
- 1. The scale of the returns problem
- 2. Fashion vs everything else: category benchmarks
- 3. RTO vs customer-initiated returns
- 4. COD: the payment mode that drives the gap
- 5. The festive-quarter spike
- 6. What one return costs
- 7. Size and fit: the #1 driver
- 8. What measurably reduces returns
- 9. The sustainability cost of returns
- 10. Structural return-avoidance: the First Resort model
- Frequently Asked Questions
1. The scale of the returns problem
Returns are e-commerce's largest hidden line item, everywhere. In the US, retailers projected USD 890 billion of merchandise coming back in 2024 — 16.9% of annual sales per NRF — while Appriss Retail and Deloitte measured USD 685 billion at 13.21% of sales, with online purchases returning at 24.52% against 8.72% in-store.[13][15] NRF expects another USD 849.9 billion back in 2025, with 19.3% of online sales returned.[14] India's all-category rate looks tamer on paper: Unicommerce's FY23 index put returns at 10.4% of orders, up from 9.8% a year earlier.[3] But that average hides a fashion category running at two to three times the all-category rate, and an RTO layer that spikes violently every festive season.[8]
2. Fashion vs everything else: category benchmarks
Attribution matters here, because the most-quoted numbers are vendor data, not census data. GoKwik — whose checkout network spans 180 million+ Indian shoppers — puts fashion and apparel returns at 25-40%, the highest of any category; ClickPost's D2C benchmarks use the same 25-40% band for fashion against 10-15% for electronics and 8-12% for FMCG; Pragma reports fashion and footwear at 30-35%.[6][9][8] The pattern is global: McKinsey's Returns Management Survey found 25% apparel e-commerce returns versus 20% overall, and Coresight measured US online apparel at 24.4%.[12][11] Fashion is structurally the most-returned thing sold online — a garment is a fit hypothesis, and hypotheses fail.
RTO rate ranges by category for Indian D2C brands — fashion and apparel (25-40%) run at two to three times the electronics and FMCG bands.[9]
| Category | RTO range (%) |
|---|---|
| Fashion & apparel | 25-40 |
| Beauty & personal care | 18-25 |
| Furniture / large goods | 15-20 |
| Electronics | 10-15 |
| FMCG / consumables | 8-12 |
Within fashion, fit-sensitivity predicts the rate: PRIME AI's category data shows dresses returned at roughly 54% and skirts at 47%, while relaxed silhouettes carry far less risk.[10] Internationally the spread is wide — Germany ~50%, the UK ~35%, the US 25-30%, India's overall online clothing rate 15-20% (with COD rejection standing in for much of what other markets book as returns).[10]
3. RTO vs customer-initiated returns: two different failures
India's returns conversation collapses two distinct events. A customer-initiated return happens after delivery: the buyer accepts the parcel, tries the garment, sends it back — the fit-and-quality problem every market has. An RTO is a failed delivery: the parcel is refused, the customer unreachable, the order cancelled mid-transit, and the box rides the network twice without ever being opened. The distinction matters because the causes differ — GoKwik's analysis attributes 60-70% of RTOs to low buying intent (customer psychology at a no-commitment checkout) and only 20-25% to genuine logistics issues.[5] Fit problems are solved with size data; RTO is solved with commitment — which is why payment mode, not product, is its biggest lever.
4. COD: the payment mode that drives the gap
Cash on delivery is the engine of Indian RTO. Across FY23, Unicommerce measured COD orders returning at 20.9% against 5.8% for prepaid — a 3.6x gap that has barely narrowed since (19.3% vs 5.6% a year earlier).[3] GoKwik's network data is starker still: COD RTO near 26%, prepaid under 2%.[6] Yet cash persists because it converts — ClickPost estimates 58-65% of Indian e-commerce orders are still cash-paid, and during Diwali 2025 COD GMV grew 35%, faster than COD volumes, as shoppers placed bigger cash orders.[9][4] Every COD checkout is an unpriced option: the customer holds the right to walk away, and the brand carries the freight both ways when they do.
COD vs prepaid return rates — 20.9% vs 5.8% across FY23, widening to 58% vs under 15% in the FY26 festive quarter.[3][1]
| Period | COD return rate (%) | Prepaid return rate (%) |
|---|---|---|
| FY23 full year | 20.9 | 5.8 |
| FY26 festive quarter | 58 | 15 |
5. The festive-quarter spike
Every force that drives returns peaks together in October-November: order volumes surge (Diwali 2025 volumes grew 24% YoY), first-time buyers flood in, and COD share of checkout climbs.[4] The result, measured across Unicommerce's 410 million+ shipment dataset: RTO hit 39.2% at the November 2025 festive peak — more than one in three orders coming back — and 58% of festive-quarter COD orders returned, against under 15% for prepaid.[1] The descent is as instructive as the spike. Brands that ran three changes at once — a prepaid incentive at checkout, pin-code-level courier routing, and address verification before dispatch — brought RTO to 25.6% by January and 21.0% by March 2026.[1][2] The festive spike, in other words, is not weather; it is an operations problem with a known fix list.
The festive RTO curve, FY26 — 39.2% at the November peak, 25.6% by January as operational changes land, 21.0% by March among optimised brands.[1]
| Month | RTO rate (%) |
|---|---|
| November 2025 (festive peak) | 39.2 |
| January 2026 | 25.6 |
| March 2026 | 21.0 |
6. What one return costs
The unit economics are brutal at fashion's price points. ClickPost puts direct logistics at Rs 150-300 per RTO parcel — and Rs 450-900 fully loaded, once the written-off acquisition spend behind the order is counted.[9] GoKwik's arithmetic lands in the same range: ~Rs 200 to deliver a Rs 1,000 order, another ~Rs 85 in reverse logistics if it bounces — a Rs 200-250 loss per event on a Rs 1,000 order.[5][6] US data shows the ceiling: Optoro estimates processing a return consumes 66% of the product's price, and Coresight counts USD 25.1 billion of pure processing cost on USD 38 billion of online apparel returns in 2023.[11] Aggregated, Pragma estimates returns cost Indian e-commerce nearly Rs 2 lakh crore a year across refunds, reverse logistics, restocking and unsellable stock.[8] A brand running 30% returns is, in effect, reshipping every third rupee of revenue.
7. Size and fit: the #1 driver
Strip out the COD layer and one cause towers over the rest. Coresight's survey of US apparel decision-makers found 53% naming size/fit the leading return reason — ahead of colour (16%) and damage (10%) — and PRIME AI reports the same 53% share of apparel returns globally, with 93% of shoppers citing sizing or fit among their return reasons; Radial puts fit and sizing at 70% of fashion returns.[11][10][16] The behavioural root is measured too: Pragma finds 67-73% of Indian customers select sizes from previous purchases or general perception rather than actual measurements — guesswork colliding with inconsistent brand sizing.[8] It is why dresses (54%) return at five times the rate of consumer electronics, and why the highest-leverage fix in fashion is not a better courier but a better answer to "which size am I?"[10][18]
"Most returns are not a courier problem. They are a promise problem — the size chart, the fabric description, the photograph all made a promise the garment had to keep. Get the promise exact and the parcel stays delivered."— Ramola Bachchan, Founder, First Resort
8. What measurably reduces returns
The fix list is short and quantified. Prepaid conversion: every order moved from cash to prepaid moves from a 20.9% to a 5.8% return probability — checkout incentives, UPI-first flows and COD verification all attack this line.[3] Verification and routing: the Unicommerce cohort that fell from 39% to 21% RTO ran address verification before dispatch and pin-code-level courier selection.[1] Speed: GoKwik measures 22% RTO on 1-2-day deliveries against 35% beyond five days, and Pragma finds 8+ day deliveries run 18-24% higher returns — intent decays while the box travels.[6][8] Fit truth: accurate, measurement-based size charts and fit tools attack the 53-70% of returns that are fit-driven.[11] At network scale the stack works: brands on GoKwik avoided Rs 130 crore of RTO losses in 2023, with fashion brands cutting COD RTOs 11% — and Invesp finds 92% of shoppers would buy again after an easy return.[7][17]
9. The sustainability cost of returns
Returns are also fashion's least-discussed emissions line. Optoro attributes up to 24 million metric tons of CO2 a year to e-commerce returns, and counted 9.5 billion pounds of returned product sent straight to landfill in 2022 — much of it clothing, the category with the highest return rate (32% average, versus 7% for electronics).[18] Returns account for roughly 25% of e-commerce's total emissions, against 7% for in-person retail, because every bounce is a second (and often third) journey plus repackaging, inspection and frequently disposal — out-of-season stock is routinely cheaper to dump than to restock.[18] The mitigation data points the same direction as the economics: keeping returned products local can improve emissions by as much as 40%, and The Interline estimates disciplined returns programmes recover 30-50% of return costs.[19] The greenest parcel remains the one that never ships twice — an argument for fixing fit and intent upstream, not just reverse logistics downstream.
10. Structural return-avoidance: the First Resort model
Most of this report describes brands managing returns after the fact. The alternative is to design the causes out, and First Resort by Ramola Bachchan runs that model as brand practice rather than as a bolt-on. Prepaid-only: with no COD, the low-intent RTO channel — 60-70% of all RTOs, per GoKwik — simply does not exist; every order starts with the commitment the data says COD lacks.[5] Made-to-order: garments are cut after the order, so there is no speculative inventory to mark down, and no over-produced stock feeding the returns-to-landfill pipeline. Fit information up front: a detailed, measurement-based XS-8XL size chart guide does the fit work before checkout, attacking the 53-70% of returns that are fit-driven at the source — helped by silhouettes that are structurally forgiving. The categories bear it out: kaftans, tunics and relaxed co-ord sets carry a fraction of the fit risk of the dresses topping PRIME AI's 54% return chart, and the logic extends across occasion wear in every size.[10] No returns department had to shrink — the returns were never generated.
"We chose prepaid and made-to-order before it was fashionable, because a garment produced for a customer who has already said yes wastes nothing — no discount rack, no warehouse of maybes, no parcel riding back and forth across the country."— Ramola Bachchan, Founder, First Resort
Frequently Asked Questions
What is the average return rate in Indian fashion e-commerce?
There is no single official number, so attribute carefully. GoKwik data puts fashion and apparel returns at 25-40%, the highest of any category; Pragma reports 30-35% for fashion and footwear against a ~17% all-category Indian average; PRIME AI benchmarks India's online clothing returns at 15-20%. Unicommerce's FY23 index measured returns at 10.4% of all Indian e-commerce orders — fashion sits well above that floor.
What is RTO (return to origin) in e-commerce?
RTO is a failed delivery: the parcel is shipped but never accepted — the customer refuses it, is unreachable, or cancels while it is in transit — so it travels back to the seller's warehouse. It is distinct from a customer-initiated return, where the buyer accepts delivery and later sends the item back. RTO is dominated by cash-on-delivery orders; GoKwik attributes 60-70% of RTOs to low buying intent rather than logistics failures.
How much higher are COD return rates than prepaid in India?
Roughly three to ten times, depending on the period measured. Across FY23, Unicommerce recorded COD returns at 20.9% versus 5.8% for prepaid — a 3.6x gap. GoKwik puts COD RTO at nearly 26% against under 2% for prepaid. In the FY26 festive quarter the gap was starkest: 58% of COD orders came back, while prepaid returns stayed under 15% (Unicommerce India D2C Report 2026).
Why do returns spike in India's festive quarter?
Three forces stack: order volumes surge (Diwali 2025 order volumes grew 24% YoY), first-time buyers enter the funnel, and COD share of checkout rises — COD GMV grew 35% during Diwali 2025. Unicommerce measured RTO at 39.2% at the November 2025 festive peak — more than one in three orders coming back — before operational fixes brought optimised brands down to 25.6% by January and 21.0% by March 2026.
What does one return or RTO actually cost a brand?
ClickPost estimates Rs 150-300 in direct logistics per RTO parcel, rising to Rs 450-900 fully loaded once written-off ad spend and handling are counted. GoKwik estimates a brand spends ~Rs 200 delivering a Rs 1,000 order and loses another ~Rs 85 in reverse logistics if it bounces. In the US, Optoro estimates processing a return costs 66% of the product's price. Pragma puts the aggregate Indian cost near Rs 2 lakh crore a year.
What is the number one reason fashion gets returned?
Size and fit, by every measurement. Coresight's survey of US apparel decision-makers found 53% cite size/fit as the leading return cause — the same 53% share PRIME AI reports for fit-driven apparel returns globally — while Radial puts fit and sizing at 70% of fashion returns. The underlying behaviour: Pragma finds 67-73% of Indian shoppers select sizes from past purchases or general perception rather than actual measurements.
Which fashion items get returned the most?
Fitted, size-sensitive silhouettes. PRIME AI's category benchmarks show dresses returned at roughly 54% — the highest of any garment type — and skirts at about 47%. Loosely-fitted categories such as kaftans, tunics and relaxed co-ord sets carry structurally lower fit risk, which is one reason resort silhouettes travel well through e-commerce.
How do Indian return rates compare with the US and Europe?
India's all-category rate is comparatively low — 10.4% of orders in FY23 per Unicommerce — but its COD-driven RTO problem is distinctive. The US returned USD 685 billion of goods in 2024, with online at 24.52%, and US online apparel returns run 24.4% (Coresight). Germany leads the world at ~50% clothing returns; the UK runs ~35%.
What actually reduces returns and RTO?
The measured levers: prepaid incentives at checkout (shifting the 20.9%-vs-5.8% COD gap), address and intent verification before dispatch, pin-code-level courier routing, faster delivery (22% RTO on 1-2-day delivery vs 35% beyond five days, per GoKwik), and honest size guidance — fit tools and accurate size charts attack the 53-70% of returns that are fit-driven. Brands on GoKwik's network avoided Rs 130 crore of RTO losses in 2023; fashion brands cut COD RTOs 11%.
Are e-commerce returns an environmental problem?
Yes, and a measurable one. Optoro attributes up to 24 million metric tons of CO2 a year to e-commerce returns and counted 9.5 billion pounds of returned product sent straight to landfill in 2022. Returns account for about 25% of e-commerce's emissions versus 7% for in-person retail (CleanHub), and clothing is the worst offender at a 32% average return rate versus 7% for electronics. Every avoided return is an avoided round trip.
Does made-to-order reduce returns?
Structurally, yes — it removes the speculative purchase from both sides. The customer commits (prepaid) before production starts, so low-intent RTO — the 60-70% of RTOs GoKwik traces to buying psychology — largely disappears, and the garment is cut to a confirmed order rather than forecast inventory, so there is no marked-down, over-produced stock feeding the returns-to-landfill pipeline. First Resort runs this model as brand practice: prepaid-only, made-to-order, with a detailed XS-8XL size chart doing the fit work up front.
Why do brands still offer COD if it drives returns?
Because it converts. Cash still accounts for roughly 58-65% of Indian e-commerce orders (ClickPost), and COD GMV grew 35% during Diwali 2025 — refusing it entirely costs a mass-market brand most of its funnel. The economics flip with order value and production model: for premium, made-to-order fashion, a bounced high-AOV parcel costs far more than the conversion COD adds — which is why prepaid-only is viable at the premium end while marketplaces manage COD with verification and incentives.
Related research: Tier-2 & Tier-3 City Fashion E-Commerce India 2026 · Made-to-Measure Indian Fashion 2026 · Plus-Size Indian Fashion 2026
Sources
- Unicommerce — India D2C Report 2026. RTO 39.2% at the Nov 2025 festive peak, easing to 25.6% by January and 21.0% by March among optimised brands; 58% of festive-quarter COD orders returned vs under 15% for prepaid; dataset of 410 million+ shipments across 6,000+ D2C brands; fashion order growth 21%; 66% of incremental volumes from non-metros. View source
- Social Samosa — Unicommerce FY26 D2C analysis. Tier II and III cities drove 66% of new D2C orders in FY26; order volumes +33% and GMV +32% YoY; RTO declined from nearly 39% (Nov 2025) to about 21% (Feb 2026); based on 400 million+ order items processed on Uniware, Apr 2024 to Feb 2026. View source
- Unicommerce — India E-commerce Index FY23 (IndianWeb2 coverage). Returns were 10.4% of total Indian e-commerce orders in FY23 vs 9.8% in FY22; COD return rate 20.9% (FY23) vs 19.3% (FY22); prepaid return rate 5.8% vs 5.6%; industry order volumes +26.2% YoY; fashion & accessories volumes +19.5%. View source
- Unicommerce — 2025 Diwali festive season report (MediaBrief coverage). Festive order volumes +24% YoY and GMV +23%; prepaid orders grew 26%; COD orders climbed 22% in volume and 35% in GMV; Tier II and III cities took ~55% of total orders. View source
- GoKwik — What Is Return To Origin (RTO) In eCommerce. Indian e-commerce average RTO rate ~20-25%, rising to nearly 40% by industry, state and pin code; national average 23.18% on 180M+ shopper data; ~Rs 200 spent to deliver a Rs 1,000 order plus ~Rs 85 reverse logistics on an RTO; 60-70% of RTOs stem from low buying intent vs 20-25% from logistics issues. View source
- GoKwik — How to Reduce RTO in e-commerce. COD orders see RTO rates of nearly 26% vs less than 2% for prepaid; fashion and apparel return rates range 25-40%, the highest of any category; orders delivered in 1-2 days show 22% RTO vs 35% for shipments delayed beyond five days; typical loss of Rs 200-250 per RTO event on a Rs 1,000 order. View source
- Free Press Journal — GoKwik RTO savings 2023. Brands on the GoKwik network saved over Rs 130 crore in RTO losses in 2023, including Rs 24 crore in reverse-logistics cost; fashion brands saved ~Rs 15 crore and recorded an 11% drop in COD RTOs. View source
- Pragma — Top Return Triggers in Indian E-commerce. Indian e-commerce returns average ~17%; fashion and footwear reach 30-35%; COD RTO rates run 20-40%; returns cost the industry nearly Rs 2 lakh crore annually; 67-73% of customers select sizes on past purchases or perception rather than measurements; deliveries taking 8+ days see 18-24% higher returns. View source
- ClickPost — RTO benchmarks for Indian D2C brands. Indian D2C brands typically run RTO rates of 20-35%, and 40%+ in fashion and other high-risk categories; category bands: fashion & apparel 25-40%, beauty 18-25%, electronics 10-15%, FMCG 8-12%, furniture 15-20%; each returned parcel costs Rs 150-300 in direct logistics and Rs 450-900 fully loaded; cash still accounts for roughly 58-65% of Indian e-commerce orders. View source
- PRIME AI — Clothing return-rate benchmarks by category and country. India overall online clothing return rate 15-20% (COD rejections often replace traditional returns); Germany ~50%, UK ~35%, US 25-30%; dresses ~54% and skirts ~47% return rates; fit issues account for 53% of apparel returns globally; 93% of shoppers cite incorrect sizing or fit as a reason for returns. View source
- Coresight Research — US online apparel returns (3DLOOK summary). Average return rate of online apparel orders in the US is 24.4%, translating to ~USD 38 billion of returns on a USD 155.8 billion online apparel and footwear market in 2023, with ~USD 25.1 billion in processing costs; size/fit cited by 53% as the leading return cause; processing a return typically costs 66% of product price (Optoro). View source
- McKinsey & Company — Returning to order (REVER summary). McKinsey Returns Management Survey noted a 25% return rate for apparel on e-commerce channels, compared to 20% overall. View source
- NRF & Happy Returns — 2024 US retail returns. US retail returns projected at USD 890 billion for 2024; retailers estimate 16.9% of annual sales returned; holiday return rates run 17% higher than the annual average; 51% of Gen Z consumers admit to bracketing. View source
- NRF — 2025 US retail returns forecast. Consumers expected to return USD 849.9 billion of merchandise in 2025 — 15.8% of annual sales; an estimated 19.3% of online sales will be returned. View source
- Appriss Retail & Deloitte — 2024 returns (Digital Commerce 360 coverage). US consumers returned USD 685 billion of merchandise in 2024 (13.21% of retail sales); online return rate 24.52% vs 8.72% in-store; USD 103 billion lost to fraudulent and abusive returns. View source
- Radial — Fashion e-commerce returns. Fashion e-commerce faces a USD 218 billion returns challenge; online clothing return rates average 30% (some reports up to 50%); poor fit and sizing account for 70% of returns in fashion retail. View source
- Invesp — E-commerce return statistics. At least 30% of products ordered online are returned vs 8.89% in brick-and-mortar stores; 67% of shoppers check the returns page before purchase; 92% would buy again if the return process is easy. View source
- CleanHub — Environmental impact of e-commerce returns (Optoro data). Up to 24 million metric tons of CO2 attributed to e-commerce returns each year (Optoro); 9.5 billion pounds of returned products sent straight to landfill in 2022; average 32% return rate for clothing vs 7% for consumer electronics; returns account for 25% of e-commerce emissions vs 7% for in-person stores. View source
- The Interline — The spiraling environmental cost of e-commerce returns. 20-30% of online purchases are returned vs ~9% in physical stores; keeping returned products local can improve emissions by as much as 40%; investing in returns change can recover 30-50% of return costs. View source