You just got your courier bill and it's higher than you expected. Not because you shipped more orders, but because a chunk of them came back. Some never reached the customer. Some reached, got opened, and got sent back. Either way, you paid twice and earned nothing.
Every D2C founder in India hits this wall. Returns feel like a customer-service problem, so you either say "no returns" and watch conversions drop, or you say "free returns, no questions asked" and watch your margin quietly bleed out. Both are wrong. The right answer is a policy that a genuine buyer reads and feels safe, while a serial abuser reads and moves on.
This guide shows you how to write that policy. We'll separate the two very different problems (RTO and customer returns), price out what a return actually costs you, and build a returns framework that is fair, legal, and margin-safe.
RTO (order comes back undelivered) and customer returns (delivered, then returned) are two different problems with two different fixes. Indian law requires a clear, visible returns/refund policy and forbids refusing defective, wrong or late goods. A single return costs you both shipping legs, packaging, a quality check, and often a write-off · roughly ₹120 to ₹250 on a mid-priced order, and reverse logistics quietly eats 12 to 18 percent of net revenue for many brands. The winning move for a thin-margin brand is exchange-first with clear condition rules: a tight return window, unused-condition proof, exchange or store credit offered before a cash refund, and refunds only to the original payment source. Generous beats stingy for trust, but generous with guardrails beats generous with none.
First, stop calling everything a "return"
Two things get lumped together and they need completely different fixes.
RTO (return to origin) is an order that never gets delivered. The customer refuses it at the door, doesn't pick up, isn't home, or gave a bad address. The courier ships it straight back to you. The customer never touched the product. This is mostly a COD problem, because a customer who paid nothing has no reason to stay home for the delivery.
A customer return is an order that did get delivered. The customer opened it, decided it didn't fit or wasn't what they wanted, and asked to send it back. This is a product, sizing and expectation problem.
The numbers make the split obvious. In India, RTO on COD orders commonly runs 25 to 40 percent and can touch higher in fashion, while prepaid RTO is often under 3 percent (GoKwik, Shadowfax). And COD is still 60 to 65 percent of Indian ecommerce orders. So your biggest return problem might not be your policy at all · it might be your payment mix.
| RTO (undelivered) | Customer return (delivered) | |
|---|---|---|
| Customer touched product? | No | Yes |
| Main driver | COD, bad address, no confirmation | Fit, colour, quality gap, buyer's remorse |
| Main fix | Push prepaid, verify COD, address checks | Better sizing, honest photos, clear policy |
| Your policy affects it? | Barely | Directly |
Your returns policy is aimed at the second column. Your RTO fix is a different job entirely · we cover it in reducing RTO on COD orders. Don't try to solve one with the other.
What a return actually costs you
Founders think a return costs "the shipping." It costs far more, and it costs it on both legs.
Take a delivered order that comes back. You've already paid forward shipping to send it. Now you pay reverse pickup to get it back. Then someone has to receive it, open it, and run a quality check (QC) · a physical inspection to decide if the item is resellable or a write-off. If it's clean, you restock it (relabel, repack, put it back in inventory). If the tag is off, it smells of perfume, or it's been worn, it's a write-off · you eat the full product cost.
So a return that comes back clean still costs you around ₹195 in pure logistics and labour. A return you can't resell costs closer to ₹395. On a product you sell for ₹799, one write-off wipes out the profit from three or four good orders.
Now zoom out. For many D2C brands, reverse logistics · pickup, QC, refunds and the support time around them · silently consumes 12 to 18 percent of net revenue (bePragma). That's not a rounding error. That's the gap between a brand that's profitable and one that isn't. If you haven't priced returns into your unit economics, your real margin is thinner than your spreadsheet says. According to the Margin Waterfall™ framework, return cost sits between RTO loss and net profit · money that's real but usually found out after the ads have spent it.
The legal baseline you can't skip
Before you get clever with policy, know the floor. The Consumer Protection (E-Commerce) Rules, 2020 apply to you the moment you sell online in India, even from your own Shopify store.
Two things are non-negotiable. First, you must clearly display your return, refund, exchange and cancellation policy before the customer buys · reasonable, transparent, and easy to find (PSL Chambers). "No returns" hidden in fine print doesn't count. Second, under Rule 4(11), you cannot refuse to take back goods that are defective, deficient, spurious (fake), the wrong item, or delivered late (India Law). A "non-returnable" tag only covers change-of-mind. It never covers a broken or wrong product.
On refunds: once you accept a return, you must refund within a reasonable period as prescribed by RBI, and it must go back to the original payment source. In practice, refunds to UPI clear in a day or two, cards in 5 to 7 working days (EnKash). Build your policy around "refund initiated within 48 hours of QC passing" so you're always inside the line.
Writing "All sales final, no returns" to protect margin. It does the opposite. It's not enforceable for defective or wrong goods under Rule 4(11), it tanks your conversion rate because nervous first-time buyers won't risk it, and it invites payment disputes and chargebacks that cost you more than a clean return would have. One founder I know ran this line for a quarter, saw checkout conversion sit 30 percent below category norm, and had no idea the policy page was the leak.
The five decisions that make or break your policy
A returns policy is really five decisions. Get these right and the wording writes itself.
1. Return window: how long?
Longer windows build trust but invite abuse and wardrobing (wearing an item, then returning it). Shorter windows are cleaner and easier to enforce. The Indian sweet spot for most D2C is 7 to 15 days from delivery. Fashion can go 7 to 10; skincare and consumables often stay at 7 for unopened only. Don't copy Amazon's 30 days · you don't have their scale to absorb the abuse.
2. Condition rules: what state must it come back in?
Be specific. "Unused, unwashed, tags intact, original packaging, no odour or stains." Vague rules lose every dispute. Take a QC photo at pickup and at receipt · you'll need it when a customer claims they returned it perfect.
3. Who pays return shipping?
If the fault is yours (wrong or defective item), you pay · the law backs the customer here anyway. If it's a change of mind, you can charge a reverse-pickup fee or deduct it from the refund. Many Indian brands offer free exchanges but paid returns. That single rule nudges people toward exchange and keeps the sale.
4. Refund or exchange or credit?
This is the big one. Offer them in order: exchange first, store credit second, cash refund last. More on this below.
5. What's non-returnable?
Innerwear, opened cosmetics, custom-made, final-sale clearance · list them clearly upfront. Remember the legal carve-out: even these get replaced if they arrive damaged or wrong.
| Category | Typical return rate | Suggested window | Return posture |
|---|---|---|---|
| Apparel / footwear | 25 to 40% | 7 to 10 days | Exchange-first, size support, paid return |
| Jewellery / accessories | 10 to 20% | 7 days | Photos-heavy, store credit preferred |
| Skincare / consumables | 2 to 8% | 7 days, unopened | Refund on defect only, sealed-only returns |
| Home / decor | 8 to 15% | 7 to 10 days | Damage claims fast, change-of-mind paid |
Apparel return rates in India run 25 to 40 percent, the highest of any category, driven almost entirely by fit and colour (industry analysis). Consumables sit at a fraction of that because size and fit don't apply. Your policy should match your category, not a template.
Why exchange-first wins for a thin-margin brand
A cash refund is the worst outcome for you and often not what the customer even wants. Most fashion returns aren't "I hate this," they're "wrong size" or "wrong colour." The customer still wants a product from you. They just want the right one.
So structure the return flow to offer, in order:
Exchange · same product, different size or colour. Revenue stays on the books. The customer's actual problem gets solved. Offer this free to make it the easy choice.
Store credit · a wallet balance, often with a small bonus (return worth ₹799, get ₹850 credit). It keeps the money inside your brand and brings the customer back for a second purchase. This is how you turn a return into a retention event.
Cash refund to original source · the last option, offered plainly so you stay trustworthy and legal, but not the default button.
Founder Decision Loop™ for returns: for every return request, resolve in this order · fix the fit (exchange) → keep the money in-house (store credit) → refund to source (cash). You only reach cash when the first two genuinely can't solve the customer's problem. This one ordering can move 40 to 60 percent of would-be refunds into retained revenue without a single unhappy customer.
Indian brands running exchange-first through a returns portal routinely keep a large share of return value in-house instead of paying it out (Base). The customer is happy because their problem got solved faster. You're happy because the cash never left.
I've watched founders treat every return as a loss to be blocked. The ones who grow treat a return as a second chance to close the sale. A returns portal that offers a one-click size swap before it ever shows the "refund" button will out-earn a stingy policy every quarter. Fight the refund, not the customer.
Cutting returns at the source
The cheapest return is the one that never happens. Most returns are avoidable, and they're avoidable before checkout, not after.
- Fix sizing. Real measurement charts, model dimensions, "fits true / size up" notes, and a size-recommender cut fit returns hard. Fit is the number-one apparel return reason in India.
- Show the product honestly. Real photos in real light, accurate colour, video on the model. The bigger the gap between the photo and the parcel, the higher your return rate.
- Set delivery expectations. Clear tracking and dates cut "I forgot I ordered this" refusals.
- Read your return reasons. If 40 percent of returns on one SKU say "colour different," that's a photography fix, not a policy fix. The data tells you where the leak is.
- Publish a clear return/refund/exchange policy on a dedicated page, linked at checkout (legal requirement).
- Set a category-appropriate window: 7 to 15 days from delivery.
- Write specific condition rules: unused, tags on, original packaging, no odour or stains.
- State plainly who pays return shipping · you on your error, customer on change of mind.
- Build the flow exchange-first, store credit second, cash refund last.
- List non-returnable items, and note the damaged/wrong-item legal exception.
- QC every return with photos at pickup and receipt before approving a refund.
- Refund to original source within 48 hours of QC passing.
- Track return reasons by SKU and fix the top three at the source.
- Flag serial returners and apply stricter terms without blocking genuine buyers.
If your margin is thin (under 30 percent contribution) → run a tight 7-day window, exchange-first, paid change-of-mind returns. If your margin is healthy and repeat rate matters more than per-order profit → widen to 15 days and free returns to lift conversion. If one SKU has a return rate above 35 percent → pause the ads on it and fix sizing or photos before you spend another rupee acquiring returns.
Your next action today
Open your current returns policy page. If you don't have one, that's your answer · write it today, because you're non-compliant and leaking conversions right now. If you do have one, run it against the five decisions above and the checklist. Then pull your last 50 returns and sort them by reason. The top reason is where you start. Fix that one thing this week and you'll cut returns before you touch a single line of policy.
Returns aren't the enemy. A vague, one-size-fits-all policy is. Get specific, go exchange-first, and price the real cost into your product pricing so a return never surprises your cash flow again. If you're still building the store, wire this into your Shopify setup and pick a courier partner from the aggregator comparison that handles reverse pickups cleanly. This is the same fulfilment thinking that runs through starting a D2C brand in India, and it's the kind of operational detail Ravikant Tyagi builds into every system.
If you'd like the complete execution system, calculators, SOPs, templates and operating frameworks behind this process, continue inside D2C Acquisition.Lab.
