If you came here wanting to set a Shopify return window by product, the short answer is that you no longer need an app for it. Since 29 September 2026, return rules in Shopify admin support overrides, so you can keep a 15-day store policy and still say 5 days on the end-of-season collection, 30 days on the gift-boxed festive range, and no returns at all on pierced jewellery. The override lives at market, collection, product and variant level, which means the discounted kurta set and the full-price one can carry different windows without you writing two policy pages or confusing anyone at checkout.
The timing is convenient. Shopify shipped this a few weeks before Diwali orders start landing, which is exactly when the refund tail gets expensive.
Where the override sits and how to set it up
Return rules live under Settings in admin, alongside your return reasons and restocking fee. The store-wide window stays where it was. What's new is that you can add a rule that applies to a narrower set of products and give it its own window, or mark that set as final sale.
Practical setup, in the order we do it:
- Fix the store default first. Most Indian D2C stores sit at 7, 10 or 15 days from delivery. Don't shorten the default to solve a festive problem — that's the policy customers compare before they buy.
- Build the collections you want to govern before you build the rules. A smart collection on tag
final-saleoreoss-octis easier to maintain than a hand-picked list of 300 SKUs. - Add the override against that collection, then test with a real order on a test product. Place it, fulfil it, and open the return request as a customer.
- Check the product page and order confirmation actually show the shorter window. If your theme hardcodes "15-day easy returns" in a static block under the add-to-cart button, the override is invisible and you've built a dispute, not a policy.
On precedence: the more specific rule is the one that applies, but a product sitting in two collections with two different overrides is the case that bites. We get this wrong on the first pass often enough that testing is non-negotiable. Decide which collection is authoritative, keep products out of the other, and re-check after your merchandising team rebuilds collections for the festive drop.
Setting a different return period for sale items
Discounted stock returns at a much higher rate than full-price stock, and it comes back later in the window. People buy three sizes at 50% off because the downside is a courier pickup. That behaviour is rational for them and brutal for you, because the item comes back in week two, misses the rest of the sale, and gets marked down again in January.
A 5-day window on sale items does two things. It compresses the decision, and it gets the unit back on the shelf while the price is still live. We generally don't recommend going to zero on apparel sale stock. Final sale on a ₹899 t-shirt is fine; final sale on a ₹4,500 discounted lehenga will cost you more in support tickets and Meta review-section damage than it saves.
The arithmetic on a shorter window
Use your own numbers, but here's the shape of it for a mid-size apparel store over a festive run.
Say 3,200 of your orders are discounted stock, AOV ₹1,850, and 22% come back. That's 704 returns. Per return you're paying roughly ₹95 for reverse pickup through Shiprocket or Delhivery, about ₹40 in QC and restocking labour, and ₹37 in gateway fees you don't get back on a prepaid Razorpay order (2% of ₹1,850). Call it ₹172 a return, so ₹1,21,088.
Then there's the stock you can't resell. If 15% of those 704 units come back marked, worn or without tags, that's 106 units written down. At a landed cost of ₹740 a unit, ₹78,440 gone. Total festive damage on discounted stock: about ₹1,99,500.
Cut the window from 15 days to 5 on that collection and suppose a quarter of those returns simply don't happen. 176 fewer returns saves ₹30,272 in handling and roughly ₹19,240 in write-downs. A touch under ₹50,000 on one festive cycle, from a settings change that takes an afternoon. Your reduction may be a third, or it may be a tenth. The point is that it's measurable and you should measure it: tag the collection, and compare return rate against the same collection last year.
Festive gifting needs a longer window, not a shorter one
This is where the override earns its keep in both directions. Someone buys a gift set on 8 October for Diwali on 1 November. Under a 10-day policy from delivery, the recipient opens a broken ceramic diya on day 22 and your support team is arguing about a policy that was never designed for gifting.
So extend it. Put your gift-boxed and festive hamper collection on 30 days, or set the window to run from a fixed date rather than from delivery if the rule allows it, and say so on the product page. A longer gifting window is one of the few customer-facing promises that costs very little, because gift recipients return at a low rate and usually want an exchange rather than a refund.
The combination is the real play: 30 days on gifting, 5 days on discounted stock, 10 days everywhere else. One store, one policy page, three windows, no contradiction.
How to exclude products from returns entirely
Mark the set as final sale in the rule rather than hiding the exclusion in policy text. Categories where we'd exclude by default in India:
- Innerwear, swimwear and socks, for hygiene.
- Pierced earrings and nose pins.
- Opened cosmetics and anything with an FSSAI-regulated food licence on the label. Packaged food comes back unsellable every time.
- Personalised and made-to-order pieces, including name-engraved jewellery and custom-stitched blouses.
- Clearance below a price floor you set — say anything under ₹499 after discount, where the reverse pickup costs more than the unit.
Two cautions. Exclusion is not a licence to refuse a damaged or wrong item, and a customer who receives the wrong SKU is entitled to a resolution regardless of what the rule says. And India's e-commerce rules expect your returns and refund terms to be displayed plainly before purchase, so a final-sale badge on the product page and a line in the cart is the minimum. Shortening a window is allowed. Burying it is the thing that gets you a consumer complaint.
Return windows for jewellery and apparel are not the same problem
Apparel returns are a fit problem. Jewellery returns are a price problem.
If you sell gold at the day's rate, a 15-day return window is an open option on the metal price. The customer buys a 22K chain at ₹6,480 a gram, the rate moves to ₹6,310 over the next fortnight, and the return arrives on day 14. You refund at the invoice value and rebuy at the market. On a 20g piece that's ₹3,400 of pure rate exposure on one order, before you count the GST adjustment and the hallmarking cost of recertifying a returned piece.
Most Indian jewellers handle this by refunding the metal at the prevailing rate on the day of return and keeping the making charge, which is fair but needs to be stated clearly on the product page. A 7-day window on gold and a 48-hour window on rate-locked pre-orders is more common than a blanket 15. If your prices move with the market, the product page has to show the breakup — metal, making, wastage, GST — so the refund logic makes sense to the buyer later. Our Gold & Silver Live Rate app does the repricing and the breakup table on the storefront, which is usually what's missing when a rate-based refund turns into an argument.
Fashion and silver-plated jewellery is a different animal. Low value, high return rate, hygiene issues on earrings. Short window, exclude the pierced lines, done.
Market-specific return policy, when you sell outside India
If you run one store across India and the UAE, or India and the US, the override at market level is the cleanest thing in this release. A 7-day window in India and a 30-day window in the US is not inconsistency, it's logistics. Reverse pickup inside India through Delhivery is cheap and fast. A return from Dubai to a Bengaluru warehouse is neither, which is why GCC-facing stores usually run a local returns address or a longer window with a restocking fee rather than free returns.
Set the market rule, then confirm the policy page served to each market reflects it. Shopify Markets will swap currency and pricing automatically; it will not rewrite the prose in your refund policy.
Native returns versus a returns app
Blunt version: if you were paying for a returns app only to control the window and show a return reason dropdown, cancel it. That's now in the platform.
Keep the app if you need any of these, because native returns doesn't cover them well:
- Exchange-first flows, where the customer picks a replacement size before the pickup is scheduled and you hold the stock.
- Store credit as the default refund option, which is how you keep margin on a return instead of losing it.
- COD refund collection. Native returns assumes you can refund the original payment method. For a COD order you need the customer's UPI ID or bank details, validated, before the pickup goes out. That whole flow is app or custom territory.
- Deep reverse-pickup integration with Shiprocket, Delhivery or Blue Dart, with automated pickup scheduling and QC photo capture at the doorstep.
- Fraud rules on serial returners, which matter more than most founders think once you cross a few thousand orders a month.
For a store doing a few hundred orders a month, mostly prepaid, native plus a WhatsApp follow-up is enough. For anything with heavy COD, the app pays for itself in refund-chasing time alone.
Reducing festive returns before the window matters
A shorter window is a backstop. The cheaper interventions sit earlier:
Address and intent confirmation on COD. A WhatsApp confirmation sent within an hour of order placement, with a one-tap convert-to-prepaid offer at 5% off, moves a meaningful slice of COD volume to UPI. Prepaid orders are refused less and returned less, because the buyer has already committed money.
Size data on the PDP, not in a popup. Garment measurements in inches, model height and the size worn, and fabric stretch. Most apparel returns in India are "size issue" because the size chart was a generic S/M/L table copied from the supplier.
Photographs that match the lighting. Colour mismatch is the second reason. Shoot on a neutral background and show the fabric close up.
Don't oversell the festive window in ads. "Easy 30-day returns" in a Meta creative is a promise your override contradicts. We've seen the chargeback that follows.
Quick answers
Can I set a return window on a single variant? Yes. Variant-level overrides are supported, which is how you make the XS and XXL of the same style final sale while the mid sizes stay returnable.
Does the override change the refund method? No. It governs eligibility and the window. Refund routing is still your payment gateway's problem, and for COD it's still a manual or app-driven bank transfer.
What about GST on a returned order? A return is a credit note against the original tax invoice. Your accounting integration needs to generate it, and the window length doesn't change the obligation. Shorter windows do keep more returns inside the same GST period, which your CA will appreciate.
Will a shorter window hurt conversion? On full-price inventory, probably. On deep-discount stock, we've not seen it move add-to-cart rate in any way worth worrying about. Price is doing the work there, not the policy.
What to do this week
Pull your return rate by collection for last festive season. If the gap between your discounted stock and your full-price stock is more than six or seven percentage points, tag the sale collection, set a 5-day override on it, extend the gifting collection to 30, and push the change before your Diwali campaigns go live. Then check the product page renders it.
If the theme work is the blocker, that's a half-day job for a developer who knows Liquid. We do this kind of thing from our Bengaluru team, and you can hire a developer by the sprint if you'd rather not take it on internally. Or send us the store for a free audit and we'll tell you where the policy text and the actual rules disagree.