Ecommerce returns cost US retailers an estimated $849.9 billion in 2025, with online return rates nearly double those of physical stores. The reasons vary sharply by category, meaning the same blended return rate can hide entirely different and fixable operational problems.
Sizing and Fit – drives 53% of apparel and footwear returns
Product Mismatch – one in five shoppers say item looked different
True Cost Per Return – ranges $43–$70 once lost LTV included
Category Benchmarks – apparel returns 3x higher than furniture
Returnless Refunds – cheaper than processing low-value items back
Blended Return Rate – conceals real causes worth fixing operationally
The most common reason customers return online orders is sizing or fit, accounting for 53% of apparel and footwear returns. The second most common is that the product looked different from what the customer expected, cited by roughly one in five online shoppers. Both stem from the same limitation of ecommerce: shoppers cannot touch, try on, or inspect products before they buy.
The scale of the problem is easy to underestimate. US retailers took back an estimated $849.9 billion in merchandise in 2025, according to the National Retail Federation. Online purchases were returned at a rate of 19.3%, compared with 8 to 10% for in-store purchases.
Yet a single blended return rate often conceals the problems worth fixing because it combines fundamentally different return causes into one metric.
Two brands can report the same return rate while facing entirely different operational problems. One may struggle with sizing, another with fraud, and a third with fulfillment errors. The overall percentage tells you how much is coming back. It does not tell you why.
This guide breaks down the eight most common return reasons by vertical, customer cohort, and acquisition channel. It also estimates the operational and financial cost of each return type so brands can identify where intervention will have the greatest impact.
Once an item comes back, it moves through receiving, inspection, repackaging, and either restocking or markdown. Those costs add up quickly. Optoro estimates the average processing cost at $46 per returned item, while Coresight Research estimates US apparel retailers spend roughly $38 billion annually handling returns. When lost customer lifetime value is included, apparel fit returns cost an estimated $43 to $70 each
| Cost component | Typical range (apparel) |
| Return shipping (outbound + inbound) | $8 to $12 |
| Processing and quality assurance | $5 to $8 |
| Markdown risk | $5 to $15 |
| Lost customer lifetime value (LTV) | $25 to $35 |
| Total per return | $43 to $70 |
Component ranges: Coresight Research and Optoro, compiled via Size AI (2026). Figures are apparel-specific and vary across product categories.
Lost customer lifetime value is often the largest contributor to the overall cost of a return. When a return becomes the customer's final interaction with a brand, the business loses not only the revenue from that order but also the future purchases that customer might have made.
Even when the returned product can be resold, that lost future value weighs on profitability more than the per-order handling cost does, because it removes earnings the brand had already counted on.
This cost structure also explains why returnless refunds have become more common. When the combined cost of shipping, processing, and restocking exceeds the value that can be recovered from the item, issuing a refund without requesting the product back may be the more economical option.
Many brands base that decision on factors such as item value, resale potential, and customer history. The trade-off is that the same policy can increase fraudulent claims, a challenge discussed later in the section on return fraud.
Around 19.3% of US online orders were returned in 2025, compared with 8 to 10% for in-store purchases (NRF Retail Returns Landscape, 2025). That headline number is close to useless for planning, though, because the variation between categories is enormous.
Return rates vary significantly across product categories. Clothing is returned at more than three times the rate of furniture, and the reasons customers return a dress have little in common with why they return a sofa. Category-level benchmarks provide a clearer picture of where returns originate and which operational problems deserve attention.
| Vertical | Return rate range | Top return reason | Est. cost per return |
| Apparel & fashion | 24 to 30% | Sizing / fit | $12 to $18 |
| Footwear | 15 to 20% | Sizing / fit | $12 to $20 |
| Accessories | 10 to 14% | Defective / appearance mismatch | $6 to $12 |
| Food & beverages | 10 to 12% | Damage / quality | $8 to $15 |
| Consumer electronics | 8 to 12% | Not as described / defective | $30 to $65 |
| Cosmetics & body care | 7 to 11% | Didn't match expectations | $6 to $12 |
| Books, media & games | 7 to 9% | Wrong item / defective | $5 to $10 |
| Furniture & household | 6 to 10% | Damage / scale mismatch | $55 to $90+ |
Return rates: estimated ranges based on Statista Consumer Insights category figures (survey of 9,778 US adults, April 2024 to March 2025), widened to reflect variation across brands and price points.
Cost per return is estimated by applying the benchmark processing cost of about 27% of item price (a $100 item costs roughly $30 to process) to representative category prices, with apparel and footwear adjusted upward for fit-return handling and furniture reflecting published LTL freight costs of $55 to $90+ per item.
These are directional estimates for processing cost, not published per-category figures, and they exclude the lost customer lifetime value discussed earlier.
The operational insight lies in the variation between categories. A retailer selling both clothing and furniture is effectively managing two different returns operations: one driven by fit and high return volume with relatively low processing costs, the other driven by freight and damage with lower return volume but substantially higher costs per return.
Averaging both into a single company-wide return rate obscures those differences and makes it harder to identify where improvement efforts will have the greatest impact. Category-level return rates provide a more meaningful benchmark because they align the metric with the underlying causes of returns.
Before the details, here is the short list operators see most often across categories:
Sizing or fit issues
Product looked different than expected
Damaged or defective on arrival
Wrong item shipped
Buyer's remorse or changed mind
Late delivery (arrived after it was needed)
Bracketing (intentional multi-buy-to-return)
Gifting returns
Nowhere is the expectation gap wider than in clothing and footwear, where a shopper is guessing at fit from a size chart. Coresight Research attributes 53% of apparel returns to size and fit, more than any other single cause.
That share lines up with the 24 to 30% clothing return band from the category table above. Size guides with real model measurements, size-specific customer reviews, and virtual try-on tools each chip away at the problem by narrowing the distance between what a shopper pictures and what shows up.
A close second is the order that simply does not match its own listing. Salsify's 2025 Consumer Research found that 71% of shoppers have returned an item because it differed from what the product page showed. This is the reason most within a merchant's direct control.
Photography, 360-degree imagery, accurate color, and honest dimension callouts are all decisions made in-house. Rebuilding a product page costs far less than absorbing the returns it prevents.
Damage concentrates in home goods, furniture, and electronics, where a single mishandled parcel can wipe out the margin on an entire order. What makes it hard to pin down is that the fault can lie with the manufacturer, the warehouse, or the carrier, and each origin points to a different owner. Packaging specs, carrier SLA audits, and proactive damage claims with 3PLs address the operational share, which NRF and other returns research place among the larger non-behavioral buckets.
Few returns irritate a customer more than one they did nothing to cause. A wrong-item shipment is fully preventable and fully on the merchant, and it bills twice: once to process the return, again to ship the correct item.
Industry fulfillment error rates run around 1 to 3%, and each mispick costs an estimated $40 to $75 to resolve, before counting the customer who does not come back. Barcode scanning at pick-and-pack, routine WMS accuracy audits, and a discrepancy flag before dispatch are the standard defenses. Treat each of these as a warehouse process failure to be traced, not a customer preference to be managed.
Some returns are set in motion before the package ever ships. SimplicityDX research found that 48% of shoppers made a recent online impulse buy and 56% of those regretted it.
The operational tell is the link to the acquisition channel. Paid social traffic from Meta and TikTok converts on impulse more readily than email or organic, so a brand leaning hard on social spend carries more remorse-driven returns folded into its blended rate. Segmenting returns by channel is the only way that cause surfaces.
An order that shows up after the birthday, the trip, or the deadline it was bought for has lost the thing that made it worth buying. In HubBox research reported by Chain Store Age, 25% of shoppers said they would return an item delivered late, climbing to 37% of millennials.
The fixes sit upstream of the return itself: accurate delivery estimates at checkout, proactive delay alerts on a branded tracking page, and WISMO deflection before frustration hardens into a return request. This is a broken promise more than a flawed product.
Buying the same item in several sizes or colors with the intent to keep one and send back the rest, a habit known as bracketing, has become routine in apparel. Younger shoppers lead it: NRF and Happy Returns report that 51% of Gen Z bracket their purchases, against 24% of Baby Boomers, and that Gen Z averages 7.7 online returns a year, more than any other generation. Because it is a deliberate habit rather than a mistake, the levers differ: fit-confidence tools, exchange incentives, and return policies segmented by customer cohort.
The final reason is seasonal and structural at once. Post-holiday gift returns create a January wave that behaves unlike everyday returns, and it is sizable: the NRF expects about 17% of holiday purchases to be returned, with the bulk landing in January once gifts are opened.
The person making the return is the recipient, not the buyer, so they often arrive with no receipt and no order number, which breaks a standard return flow. Gift receipts in the box, a dedicated gift-mode return path, and clearly communicated extended windows keep the January spike from overwhelming support.
Some returns have nothing to do with a product falling short. Signifyd's 2025 State of Fraud and Returns report found that abusive returns rose 64% between January 2024 and May 2025, with their overall share nearly doubling. A blended return rate absorbs this alongside legitimate returns, so it stays invisible until a brand separates it out.
The most familiar version is wardrobing: buying an item, using it once, then returning it inside the window, whether that is a dress worn to one event or a laptop bought for a single deadline. It clusters in fashion and consumer electronics. NRF's 2025 report estimates that 9% of all returns are fraudulent, and other industry estimates run higher depending on how abuse is defined.
Bracketing shades into abuse when it becomes deliberate policy-gaming, and the fix is a real bind. Tighten the return window and cart abandonment rises; loosen it and abuse climbs.
What works is not a single setting but segmented rules that treat a loyal customer differently from a serial returner. Newer still is agentic AI fraud, where automated scripts and shopping agents file return claims at scale and probe returnless-refund thresholds, a pattern Signifyd flags as emerging in 2025.
Return data creates value only when it leads to action. ClickPost organizes its returns capabilities around the same three operational problems covered in this guide: expectation gaps, fulfillment and delivery issues, and policy or abuse.
Expectation gaps: ClickPost's returns intelligence captures return reasons at the SKU level, helping merchandising teams identify products with recurring fit or appearance-related issues and prioritize product page improvements. The returns portal also uses the selected return reason to drive AI-driven nudges toward exchanges instead of refunds where appropriate, helping brands retain revenue.
Fulfillment and delivery: Branded tracking pages and milestone-based notifications keep customers informed throughout the return and refund journey, reducing WISMO inquiries and support workload. The returns portal supports site-wide and variant exchanges, SKU-level eligibility rules, instant refunds to store credit or gift cards, and map-based PUDO drop-off selection within a fully customizable experience.
Policy and abuse: ClickPost's customer segmentation module groups shoppers based on their return history and configurable thresholds, allowing brands to tailor return policies by customer segment. Trusted customers can receive faster refunds or additional flexibility, while frequent returners and low-value-order returners can be subject to stricter rules. ClickPost does not detect fraud directly, but it gives brands the controls to limit the patterns that repeat abuse relies on. Return protection and green returns provide additional flexibility where appropriate.
ClickPost also integrates natively with Shopify, making it easier for growing D2C brands to deploy and manage their returns operations without extensive implementation effort.
This is a diagnostic sequence for a head of ecommerce or an operations manager, not generic advice. Worked in order, it turns a vague blended rate into a set of assignable problems.
1. Pull your return reason report by SKU and category.
Separate clothing, electronics, furniture, and your other major product categories. Category-level reporting makes it easier to identify the products and issues driving returns and reveals patterns that disappear in a company-wide average.
2. Cross-reference return reasons against acquisition channel.
Compare return patterns across paid social, search, email, affiliate, and organic channels. If one channel consistently produces higher return rates or different return reasons, review whether campaign messaging, audience targeting, or product expectations are contributing.
3. Segment by customer cohort.
Analyze new and repeat customers separately. Higher return rates among first-time buyers often indicate expectation gaps in product pages, ad creative, or merchandising. Rising returns among repeat customers are more likely to point to product quality or fulfillment issues.
4. Classify every return into one of three fix buckets.
Assign each return reason to the team best positioned to solve it:
Product experience: Expectation gaps, sizing, or inaccurate product information.
Fulfillment: Wrong items, damaged products, or shipping errors.
Policy or customer behavior: Bracketing, wardrobing, or return abuse.
Each bucket should have a clear owner and a corresponding corrective action.
5. Set a 90-day reduction target per bucket.
Assign an owner, define a measurable target, and track progress separately for each category. Measuring every bucket independently makes it easier to see which changes are working and prevents improvements in one area from masking problems in another.
Returns are not one problem. A single blended rate hides sizing failures, fulfillment errors, policy abuse, and impulse returns tied to channel mix, and each of those needs a different owner and a different fix. Treating them as one number is why a brand can work hard on returns and watch the rate barely move.
The scale of returns flowing back through US retail, approaching $850 billion a year, makes them too expensive to treat as a logistics afterthought. Read properly, return-reason data is a running diagnostic on your product pages, your warehouse, and your policy.
As bracketing becomes standard behavior among younger shoppers and new pressures like automated return abuse emerge, brands that build return-reason intelligence into their operations are better positioned than those still watching a single number. If you want to act on that data at the platform level, ClickPost's Returns solution is built for exactly this.
Sizing and fit are the most common reasons. Coresight Research attributes 53% of apparel returns to size and fit, ahead of every other cause. The next most common reason is that the product looked different from its listing: Salsify found 71% of shoppers have returned an item for that reason.
About 19.3% of US online orders were returned in 2025, compared with 8 to 10% for in-store purchases (NRF Retail Returns Landscape, 2025). Return rates vary widely by category, from roughly 24 to 30% for clothing down to 6 to 10% for furniture and household goods.
Optoro estimates processing a return costs about 27% of the item's price, or roughly $30 on a $100 order. For apparel, once lost customer lifetime value is included, the all-in cost of a fit-related return runs an estimated $43 to $70. Oversized items cost more because of freight.
Bracketing is buying multiple sizes, colors, or styles of the same item with the intent to keep one and return the rest. It is most common in apparel and skews younger: NRF and Happy Returns report that 51% of Gen Z shoppers bracket their purchases, against 24% of Baby Boomers.
Wardrobing is buying an item, using it once (wearing it to an event, using electronics for a project), then returning it within the window. It is a form of return abuse most common in fashion and consumer electronics.
Returns erode margin well beyond the refunded amount. When a return is a customer's last interaction with a brand, the lost future value of that customer often outweighs the per-order handling cost, because it removes revenue the brand had already counted on.
Customers cannot physically inspect, try on, or test products before buying online, so the gap between expectation and reality is wider. That pushes online return rates roughly two to three times above in-store rates.
Start by segmenting return reasons by category, acquisition channel, and customer cohort. Then apply targeted fixes: rebuild product pages for expectation-gap returns, audit fulfillment for operational errors, and adjust policy for behavioral returns like bracketing and abuse.