Key Takeaways
- The refund fight leaves the building. Of coded responses to a refused return, about 64% are chargebacks or card disputes, against 10% who complain publicly and 12% who keep arguing with support. Most of this never reaches your support queue as a complaint
- The fee is not the problem, the surprise is. Restocking fees are 37% of coded fee mentions and return shipping deducted from the refund is 34%, and the anger in both concentrates on charges that were not visible at the point of purchase
- Free returns are now the baseline, not the perk. About 26% of coded fee mentions treat them as a filter on where to shop at all, not as a reason to prefer one brand over another
- Store credit is the most resented outcome in the study at 68% of coded refund-form mentions. It reads as the brand keeping the money, and it is a trigger for the chargeback rather than an end to the dispute
- The consequence is lopsided. About 49% of coded consequence mentions are people saying they will not buy again and 38% are people warning others off, against 3% who name the return policy as the reason they keep buying
- Convenience beats generosity. In-store or drop-off returns are 46% of coded friction mentions, and the appeal is simply that nothing has to be printed, packed or posted
Returns get discussed inside most brands as a margin line. Fees exist to recover cost and to discourage the shoppers who order five things and keep one. That framing is reasonable and it is also incomplete, because it treats the return as the end of a transaction rather than the middle of a relationship.
This is a voice-of-customer study of more than 7,500 real comments about return fees, refunds, restocking charges and return policies. The corpus comes from the communities where people compare notes on buying things without a brand in the room: apparel and plus-size fashion communities, beauty and skincare communities, outdoor and durable-goods communities, general shopping and frugality spaces, and consumer-advice and personal-finance discussions. What comes through is that the policy is rarely read before the purchase and almost always discovered during the return, which is the worst possible moment for a surprise.
In a voice-of-customer analysis of 7,500+ comments about return fees and refunds, roughly 64% of coded responses to a refused or blocked return involved a chargeback or card dispute, against about 12% who kept arguing with support and 10% who complained publicly. Store credit was the most resented refund outcome at 68% of coded refund-form mentions. About 49% of coded consequence mentions were people saying they would not buy from the brand again, while only about 3% named a return policy as the reason they keep buying.
About this study
- Corpus
- 7,500+ public comments where people discuss return fees, restocking charges, return shipping, refunds, store credit and what they did when a return was refused
- Communities
- Apparel and plus-size fashion communities, beauty and skincare communities, outdoor and durable-goods communities, general shopping and frugality spaces, and consumer-advice and personal-finance discussions
- Themes coded
- What the fee actually costs you, the window and the errand, the refund that is not money, how people respond when a return is refused, and what the experience costs the brand
- Analysis
- Directional thematic coding by keyword and pattern, applied within topically matching threads. A comment can be coded into more than one theme, so percentages are the share of coded mentions within a section, not of the full dataset. Figures are directional estimates from this corpus, not a precise census.
- Quotes
- Verbatim. Only character-encoding artifacts cleaned up. The wording, spelling, and typos are the commenter's own.
Three notes on scope. First, threads were gathered by comment volume, which favours large and active discussions, so this corpus over-represents the returns people felt strongly enough to argue about. Quiet, uneventful returns are underrepresented by construction. Second, a few of these communities include retail staff as well as shoppers, and the two groups describe the same policy from opposite sides. Quotes below are shopper-side unless the context makes the role obvious. Third, one question this study set out to answer did not survive the data. Instant or near-instant refunds are mentioned in well under 1% of the corpus, too rarely to rank or to draw a conclusion from, so no section claims a finding about them. That absence is discussed at the end.
The research question
When do return fees, slow refunds and restrictive policies stop a purchase or permanently cost a brand the customer, and which return conveniences actually rebuild loyalty? The short version: fees rarely stop the first purchase because almost nobody reads the policy first, the damage lands at the return itself, and the response is financial rather than conversational. Shoppers do not escalate to you. They escalate past you.
1. The fee is not the problem, the surprise is
Four reactions, ranked by share of coded fee mentionsReturn fees generate a lot of anger in this corpus, but reading the comments closely, the anger is not really aimed at the existence of a charge. It is aimed at finding out about the charge too late to do anything about it. The clearest evidence is the small but consistent group who say plainly that a disclosed fee is fine.
The restocking complaint is loudest when the item was never used, because at that point the shopper cannot construct a story where the fee is covering a real cost. What they construct instead is a story about bad faith.
"It’s absolute bullshit. I’m not here to keep your business afloat. I just bought and returned a pair of brand new sneakers from Moda3 online using Affirm. They mention in the fine print that they “sometimes” charge a restocking fee but they don’t say what it is or why. I never wore the shoes, returned with OG packaging…and paid 23 bucks at UPS to return. If you aren’t transparent about it I feel like it’s illegal."
"A 100 % restocking fee is diabolical and has to be illegal somewhere. At that point it’s basically a no return/ no refund policy."
The return-shipping complaint has a sharper, more specific version that is worth reading twice, because it is not about the money at all. It is about being denied a choice the shopper knows they could have made more cheaply.
"I don't mind having to pay for return shipping but I do mind it being an automatic process/predetermined method deducted from my refund when I KNOW I could have returned it more cheaply using another shipping method.They recently charged me $8 for UPS to return a tiny network adapter. There was no other option available and I know I could have done it via USPS a lot more cheaply.That sticks in my craw."
Free returns show up less as a delight and more as a qualifying criterion. The comment below is the mechanism a lot of brands miss: free returns are what makes someone willing to buy two sizes from a brand whose fit they do not know yet.
"Whenever I shop online, I’ll only purchase from stores that provide free returns (bonus points if they provide the return label with the original shipment). This way, I feel more free to purchase additional sizes if I’m unsure, or if I’m unfamiliar with the brand."
And the group that proves the fee is not the real issue. This person paid nothing in the end, and still describes the interaction as something done to them.
"Verizon rep was "showing" me how that stupid Hum thing works and activated it. When I told him I didn't want it they tried to pull the "might as well take it there is a $50 restocking fee". It took three objections and an escalation, but they finally took it back. Fees like this are pushed on people that would rather pay than have the awkward confrontation."
The commercial question is not whether to charge a return fee. It is whether the fee is visible before the card is charged, because a disclosed fee is a business term and an undisclosed one is a betrayal. Put the exact figure on the product page and in the confirmation email, not in a policy page nobody opens. The second lesson is in that shipping quote: when you force a single carrier and deduct the cost, you convert a cost into a grievance, because the shopper knows the cheaper option existed. Offering a choice of return method, even a slower one, removes the sense of being charged for someone else's convenience. And note what free returns are actually buying you. They are not a loyalty perk, they are the thing that makes a first purchase from an unfamiliar brand possible at all, which matters most if your fit or finish is hard to judge from a photo.
2. The window and the errand
Four friction points, ranked by share of coded return-friction mentionsOnce someone decides to return something, the experience is judged on effort rather than on generosity. The single most positive pattern in this corpus is not a long window or a full refund. It is being able to hand the item to someone at a place they were already going.
The best comment on return convenience is also a warning about how brittle it is. This is someone who wants to do the easy thing and is being routed into the hard thing, and their solution is to lie to the returns form.
"I basically have to lie every time I want to do a return to get it to be "free". They always offer me free UPS store drop off no matter my return reason but the closest one is 45 minutes away. I have to select one of the return reasons like it doesn't fit or is defective to be able to return it locally at my UPS authorized shipping center. I'm so over it. Like, there is a Kohl's, why can't I do my returns there?"
The fourth category is the quietest and the most expensive, because nothing happens. The shopper keeps an item they did not want, says nothing to the brand, and draws a conclusion anyway. No support ticket is ever filed, so from the inside this looks like a successful order.
Convenience is worth more than generosity here, and it is cheaper. A 60 day window that requires printing a label, sourcing a box and driving to a carrier will lose to a 30 day window with a QR code and a drop-off point. If you sell through any physical footprint or partner network, returns are one of the few places that footprint produces a measurable trust advantage, so make it the default rather than a hidden option. The subtler point is about routing logic: when your form only offers the convenient method for certain return reasons, you are training customers to misreport why they are returning, which corrupts the return-reason data your merchandising team relies on. And the silent group matters most. Someone who keeps an unwanted item because returning it is not worth the effort never appears in your complaint data, but they have already decided how much risk your brand is worth next time.
3. The refund that is not money
Three outcomes, ranked by share of coded refund-form mentionsIf there is one policy in this study that reliably converts an ordinary return into a dispute, it is refunding in something other than money. Store credit dominates the theme, and the language around it is noticeably harsher than the language about fees.
The clearest example in the corpus combines a partial refund with store credit, which is the worst of both. The arithmetic is what makes it memorable, and the last line is what makes it costly.
"I bought a dress, tried it on, and it doesn't fit for me, after 30 seconds, I put it back in the original packaging as well. The tags were still attached to the clothing exactly as they were when I received it, and I did not damage or alter the item in any way. So I returned it on time, $35.71 was the purchase price and they said there's 2 strands of hair, gave me credit for $5.36. That means they keep the dress and $30. I was a regular customer but now no shein no more..."
A second pattern sits underneath this one. A refund in store credit assumes a future purchase, which is precisely the thing the shopper is now reconsidering. It answers a question about money with an answer about loyalty, and the shopper reads the substitution as an admission.
"RealReal. They sent me the wrong item, made me send pictures via their app chat to prove it was the wrong item and then charged me their return shipping fee. Their chat doesn’t save the history and after multiple attempts and conversation to have them right their wrong I deleted their app. I spent a fair amount with them and it makes me happy to take my business elsewhere where I’m treated better"
Store credit looks like a retention tool on a spreadsheet and behaves like a retention risk in the comments. It converts a closed transaction into an open grievance, and it is the most common step immediately before someone describes calling their bank. If you use it, the honest version is to make it optional and to price it: full value as credit, or the cash refund minus a stated fee, chosen by the customer. The choice is what removes the sense of confiscation. Partial refunds deserve a harder look still. Deducting value for a condition issue may be defensible per item, but the shopper will publish the arithmetic, and "$35.71 became $5.36" travels much further than any explanation you attach to it. If you are going to reduce a refund, the reason has to be specific, evidenced and sent before the money moves, not discovered by the customer on a statement.
4. When the return is refused, people go around you
Five responses, ranked by share of coded escalation mentionsThis is the finding that reframes the whole category. When a return is refused, delayed past patience, or refunded in a way the shopper rejects, the dominant response is not to complain harder. It is to retrieve the money through a third party the brand cannot overrule.
What makes the chargeback route so durable is that it is passed between strangers as reliable advice, with the issuer named and the timeline attached. It is not a threat made in anger, it is a procedure.
"Chase is really good about these disputed claims, I bought a car radio/Navigation system from Alibaba and after the thing broke within the first week and the seller wouldn't help, I disputed the claim with Chase and I had a refund in less than a 2-3 weeks."
The public-complaint route is smaller, and one comment explains why it may be smaller than it looks. When someone tries the public route and finds it closed, they do not stop being angry. They stop being visible, and they take the volume with them.
"I bought a nespresso aerrochino 3 that was advertised as brand new... it was not. the product was refurbished. I returned it and left a scathing review to warn others, but amazon deleted it. I decided to sort reviews by newest and some reviews said the same thing. I went back a few days later and those were also deleted. I hardly ever buy from Amazon anymore, especially if I want something new or name brand."
If most refused returns end in a chargeback, then your returns policy is not really being enforced by your policy. It is being adjudicated by card networks, on their evidence standards, with fees and a dispute ratio attached. A denial that saves the cost of one item and produces a chargeback has cost you the item, the fee, the customer and a mark against your merchant account. That changes the maths on marginal denials considerably: the cheapest resolution is usually the one you grant before it leaves your control. There is also a measurement trap here. Because only about 10% complain publicly and 12% keep arguing, your support tickets and review scores are a heavily filtered view of returns anger, and the majority signal is sitting in your chargeback data instead. If those two datasets are owned by different teams, and in most brands they are, nobody is looking at the whole picture. Reconciling dispute reasons against return reasons monthly is the single highest-value thing a small team can do here.
5. What the experience actually costs
Four consequences, ranked by share of coded consequence mentionsThe consequence theme is the most lopsided ranking in the study, and the shape of it is the argument for treating returns as a retention channel rather than a cost centre.
That 3% is small, and it is also the most commercially interesting group in the corpus, because of how absolute they are. These are not people who mildly appreciate a good policy. They describe it as the entire reason the brand gets their money.
"The only reason I buy things from them is because of their return policy. I can get any item they sell from most other companies."
"One of the main reasons I shop at REI (and spend more) is the no hassle return policy. I don’t return old shoes 😆 or heavily used items but regardless - if the policy changes I won’t have a reason to go with REI."
"In my opinion, part of the price of goods at these stores is the return policy/guarantee. I will specifically purchase some items at a store with a good return policy if I am concerned about the quality or durability of the product. I am willing to pay more for a product if it has a guarantee attached to it. You're not taking advantage of these stores, you are using a service that you paid for."
The last of those is the most useful sentence in the whole corpus for anyone pricing a returns programme. The shopper is not describing a concession. They are describing something they believe they already bought, which is why withdrawing it later reads as a broken deal rather than a policy update.
On the other side, the churn statement is usually attached to a specific, small, checkable detail rather than to a general feeling. This is what makes it credible and repeatable.
"Always check the return policy for ANYWHERE you buy something. A lot of places are changing their policies. And I get it. I just had an experience with five below. Even though my item was well within the return policy. That is atrocious. And won’t be shopping there anymore. They just lost a customer."
Read the asymmetry carefully before drawing the obvious conclusion. A bad return experience is volunteered constantly and a good one almost never, which means return policy is a much stronger repellent than magnet, and the return on making a good policy slightly better is low. The return on removing a specific failure is high. The practical move is to find the two or three moments in your own returns flow that generate the language in section 4 and fix those, rather than to advertise generosity. But the 3% carry a second lesson that matters if you already have a strong policy: they treat it as part of the price they paid, so tightening it later will not be received as a cost adjustment. It will be received as a bait and switch, and this corpus contains people describing exactly that transition at brands whose policies changed.
The question this corpus could not answer
Qualitative. Reported because the absence is itself informativeThis study set out to test, among other things, how much shoppers value instant or near-instant refunds. The honest answer is that this corpus cannot say, because they barely come up. Across 7,500+ comments, explicit mentions of instant refunds and of refund speed together account for well under 1%, which is far too thin to rank or to build a recommendation on.
That is worth reporting rather than quietly dropping, because refund speed is a common feature investment. What the comments talk about instead, constantly, is refund certainty: whether the money is coming back at all, in what form, and whether they will have to fight for it. Speed appears to be a distant concern next to that. I would not conclude from this that fast refunds do not matter, only that in the communities where returns get argued about, nobody is asking for them, and a brand deciding between faster refunds and more certain ones has a clear signal about which anxiety is live.
The same caution applies to pre-purchase policy checking, which the study also set out to measure. It appears in roughly 1% of comments, too rarely to rank. The pattern in those comments is consistent though: the people who check are people who were burned before, and they use the policy as a filter that removes unfamiliar brands entirely.
"Yeah I wanted to order from them and didn’t because of the return policy. Why would I ever take a chance on a new store when I have no idea how the fit and quality is?"
For a new or unfamiliar brand, the return policy is doing work at the top of the funnel that most teams attribute to price or creative. The shopper above did not abandon a cart because of cost or copy. They declined to take a risk that the policy did not cover, and no one in your analytics will ever tell you that is why. If your fit, colour or finish is hard to judge from a photograph, the returns terms are part of your conversion rate whether or not you treat them that way, and they belong on the product page rather than in a footer link.
A customer glossary
The words people actually useIf you work on ecommerce, retention or customer support, these are the terms already in circulation among the shoppers you are selling to.
| Term | What people mean | Signal |
|---|---|---|
| Restocking fee | Treated as bad faith when the item is unopened, because the shopper cannot see what cost it covers. The phrase reliably appears alongside accusations of dishonesty rather than of expense. | complaint |
| Chargeback | Not a threat, a procedure. Shared between strangers with the card issuer named and a timeline attached, as the reliable way to settle a refused return. | behaviour |
| Store credit | Understood as the brand keeping the money. The most resented refund outcome in the corpus, and a common last step before a dispute is opened. | complaint |
| Final sale | Read as "no returns" regardless of the reason for the return, including brand error. Appears most often in complaints about clearance and outlet purchases. | objection |
| The fine print | Where shoppers say fees live, used as an accusation rather than a description. Its presence in a comment almost always precedes a dispute or a churn statement. | complaint |
| No hassle returns | The positive standard, and notably about effort rather than money. Used by the small group who name returns as their reason for buying. | benefit |
| Drop off | The preferred return method, defined by what it removes: no printing, no box, no dedicated trip. Named partner locations are cited as a reason to buy. | benefit |
| Return window | Discussed as a countdown that starts without warning. Complaints cluster on discovering the window at the moment it has already closed. | objection |
| They just lost a customer | The closing line of the churn statement, delivered as a verdict. Almost always attached to one small, specific, checkable detail rather than a general grievance. | complaint |
| Part of the price | How the loyal minority frame a return guarantee: something already paid for, not a concession granted. Explains why tightening a policy reads as a broken deal. | stance |
Frequently asked questions
What do shoppers do when a brand refuses their return?
They go around you. Of the coded responses to a refused or blocked return, about 64% involve a chargeback or a dispute through the card issuer. Roughly 12% describe arguing with support until the decision is overturned, 11% describe formal escalation to a regulator, small claims court or a lawyer, and 10% describe a public complaint such as a review or a social post. Only about 4% describe giving up and absorbing the loss. The customer service queue is not where this fight happens. The bank is.
Do return fees stop people from buying?
The fee itself is not usually the problem. The surprise is. Of coded fee mentions, about 37% are restocking fees and 34% are return shipping deducted from the refund, and the anger in both cases is concentrated on charges that were not clear at the point of purchase. A separate 26% treat free returns as the baseline expectation rather than a perk, and several describe filtering their shopping to stores that offer them. The complaint is rarely that a fee exists. It is that the fee appeared after the money was already spent.
Does store credit work instead of a refund?
It is the single most resented refund outcome in the corpus. Store credit or a gift card accounts for about 68% of coded mentions in the refund-form theme, well ahead of exchange-only and final-sale policies at 24%. Shoppers describe it as the brand keeping their money, and it is a common trigger for the chargeback route rather than a resolution that ends the dispute. Partial credit is worse still: one shopper described a $35.71 dress refunded as $5.36 in credit, and ended the comment by saying they were done with the brand.
Does a good return policy actually build loyalty?
For a small group it is decisive, but it is a far stronger repellent than magnet. In the coded consequence theme, about 49% of mentions are people saying they will not buy from the brand again and 38% are people warning others off. Only about 3% name the return policy as the reason they keep buying. The asymmetry is the finding: a bad return experience is volunteered constantly, a good one almost never. But when people do volunteer it, they are absolute about it, with one shopper saying the policy is the only reason they buy from that retailer at all.
What return method do online shoppers prefer?
Dropping it off somewhere they were already going. About 46% of coded mentions in the return-friction theme are about in-store or drop-off returns, and the appeal is that it removes packing and printing entirely. Around 20% describe the opposite experience as the friction that defines the return: printing a label, finding the original box, and making a trip to a carrier. Roughly 26% concern the return window being too short or already missed, and about 8% describe abandoning the return because the effort exceeded the value of the item.
Do shoppers check the return policy before they buy?
Less often than the industry assumes, and that is the uncomfortable part. Explicit pre-purchase policy checking is rare in this corpus, appearing in roughly 1% of comments, so it cannot be ranked reliably. What the comments show instead is that the policy is usually discovered after the purchase, during the return, which is exactly when a surprise becomes a dispute. The shoppers who do check tend to be people already burned once, and they describe using the policy as a filter that removes unfamiliar brands from consideration entirely.
Want this run for your brand or category?
This is a public sample of how we work. Insightios reads Reddit, Amazon reviews, YouTube, and the communities where your buyers actually talk, then delivers a report with the exact language, objections, and use cases behind your product.
This report analyzes consumer language and perceptions about return policies, fees and refunds. Statements about the policies or practices of any named company are reported as claims made by commenters and were not independently verified. Brand and product names appear only as they were mentioned by commenters. Nothing here is legal advice about consumer rights, which vary by country and by contract. Percentages are directional estimates from this corpus, not a census of shopper behaviour.