Cut Apparel Returns by Targeting the 53% Fit Problem for Retailers

Cut Apparel Returns by Targeting the 53% Fit Problem for Retailers

Poor fit drives the majority of apparel returns, with industry surveys placing size and fit issues anywhere from 53% to 70% of the total. Behind that, the next four culprits are product mismatch with photos or descriptions, damage or defects, changed minds and bracketing, and fulfillment errors. Fix fit first — everything else is a smaller lever by comparison.

TL;DR:

• The majority of apparel returns are caused by poor fit, with size and fit issues accounting for up to 70% of returns in some surveys.
• Improving measurement accuracy, virtual try-on tools, and size recommendations can significantly reduce fit-related returns, which are the largest category.
• Return reasons linked to product mismatch or damage require better photography, quality control, and supplier standards to cut down non-fit returns.
• Seasonal spikes, trend volatility, and specific categories like dresses and bottoms tend to have higher return rates, necessitating targeted monitoring.
• Differentiating online and in-store return causes is crucial, with online returns heavily influenced by fit uncertainty and virtual try-on solutions being particularly effective.

div style="border-radius:26px;padding:22px;background:radial-gradient(circle at 100% 0%,#fbfaf7 0 150px,rgba(255,255,255,0) 151px),radial-gradient(circle at 0% 100%,#fbfaf7 0 130px,rgba(255,255,255,0) 131px),linear-gradient(180deg,#fcfbfa 0%,#fefdfc 100%);">

div style="background:#ffffff;border-radius:18px;box-shadow:0 12px 34px rgba(0,0,0,0.13);overflow:hidden;">

div style="display:flex;flex-wrap:wrap;background:linear-gradient(104deg,#121211 0%,#121211 33%,#ffffff 33.15%);">

div style="flex:0 0 30%;min-width:150px;padding:30px 10px 30px 26px;color:#ffffff;">

div style="margin:0 0 14px;"> Garmcheck

garmcheck.com

div style="font-size:23px;font-weight:800;line-height:1.2;letter-spacing:-0.01em;color:#1f2937;margin:0;">Show Customers How Garments Fit

GarmCheck helps Shopify fashion brands address fit uncertainty with AI virtual try-on and size recommendations from eight body measurements.

See how GarmCheck works

Table of Contents

• Apparel return reasons ranked by prevalence and cost
• How to capture and analyse return-reason data effectively
• High-impact tactics that actually move the return rate
• Example: AI-driven virtual try-on tackling fit-driven returns
• Measuring success: KPIs, timelines and a simple ROI check
• Seasonal or trend-driven return reasons
• Return rates variation by apparel category
• Differences in return reasons between online and in-store purchases
• Influence of size guides and fitting technology on return reasons
• Author perspective: sequencing the levers without hurting conversion
• Where GarmCheck fits into a returns-reduction stack
• Sources

Apparel return reasons ranked by prevalence and cost

Return data across most apparel retailers tends to cluster around the same five buckets, though the order can shift by category and season. Here is how they typically stack up, along with what to do about each.

• Wrong size or poor fit. This is the dominant driver everywhere it gets measured. Coresight Research found 53% of apparel sellers name size and fit as the top return reason , and a more recent survey of 1,882 online shoppers identified incorrect sizing as the primary driver behind clothing return rates as high as 60% in that sample. The fix isn’t a single tool. It’s a stack: accurate measurement tables on every product page, a size recommender that uses body data rather than guesswork, and virtual try-on that shows the shopper what the garment will actually look like on them before they click “buy”. What to measure next: return rate by SKU×size, and whether returns cluster at specific sizes rather than spreading evenly.
• Product looks different to photos or description. Shoppers return items that don’t match what they saw online, whether that’s colour rendering, fabric weight, or drape. Better photography can reduce returns by showing multiple angles, video, and honest material notes including weight and stretch, along with real customer photos alongside studio shots. What to measure next: return rate tagged “not as described” against the specific PDP assets live at time of sale — a page redesign should show up in that number within weeks.
• Damaged or defective items. These are the returns that should never have shipped. Tightening incoming quality control checkpoints, spot-checking supplier batches, and setting a clear defect threshold before goods reach the warehouse floor all reduce this bucket. What to measure next: defect rate by supplier and by production batch, not just by SKU — a single bad batch can distort your whole category’s numbers.
• Changed mind, impulse purchases, and bracketing. Bracketing (ordering multiple sizes or colours with the intention of returning most of them) inflates return rates without necessarily indicating genuine product faults. Exchange-first return flows and policies encouraging exchanges over refunds can reduce net refund volume while maintaining customer goodwill. What to measure next: the ratio of exchanges to refunds by customer cohort, and repeat-bracketing rates among your highest-value shoppers.
• Wrong item shipped or delivery issues. Fulfilment errors are rarer than fit problems but expensive to ignore, since they usually trigger a refund and a re-ship. Barcode verification at pack stage and clear courier service-level agreements catch most of this before it reaches the customer. What to measure next: mis-ship rate by warehouse zone or picker, which usually points to a training or layout problem rather than a systemic one.

How to capture and analyse return-reason data effectively

Most brands under-invest in the taxonomy behind their return reasons, which makes root-cause work almost impossible later. A single-tier reason code (“didn’t fit”) tells you nothing about why it didn’t fit. A two-tier structure does: primary reason (“size”) plus secondary reason (“ran small in the shoulders”, “true to size but customer ordered up”). That second layer is what turns returns reason analytics from a reporting exercise into an operational tool.

At minimum, your return intake should capture:

• SKU and size ordered
• Colour or variant
• Photo evidence for damage or defect claims
• Condition grade on arrival back at the warehouse
• Free-text customer notes, tagged against a secondary reason code

Once that data exists, three routines do most of the diagnostic work. Compute return rate by SKU×size to spot which specific sizes within a style are running hot. Flag outlier sizes that return at two or three times the rate of neighbouring sizes in the same style, which usually points to a chart error rather than a broader fit problem. Run cohort analysis on repeat returners to separate genuine bracketing behaviour from a customer who simply can’t find their size.

Pro Tip: Before you touch a global size chart, check whether the problem sits in one SKU across several sizes or one size across several SKUs — the first is a manufacturing issue, the second is a chart issue, and confusing them wastes a quarter of testing.

Weekly dashboards catch emerging spikes early; monthly deep dives are where you actually rewrite size charts or renegotiate with a supplier.

High-impact tactics that actually move the return rate

Not every lever is worth the same effort. Fit-first interventions carry the largest upside because they attack the biggest bucket directly.

• Size tech and virtual try-on. Size recommenders and virtual try-on tools built on body measurement typically show meaningful reduction in fit-related returns when layered onto a PDP that already has decent measurement data.
• Single-brand size models. Brands that fit their size charts to their own actual customer measurements, rather than a generic industry standard, see fewer edge-case returns at the size extremes.
• PDP and content fixes. Measurement tables, real model data (height, size worn), diverse body-type photography, and video consistently reduce “didn’t match expectations” returns, and product presentation is one of the lowest-cost, highest-impact tests available before any technology spend.
• Operations and supplier levers. Incoming QC and a clear inspection standard reduce defect-driven returns, while a refurbishment workflow recovers resale value from items that come back in saleable condition.

Fit-related returns are consistently cited as the largest single category in apparel , which is why size and fit interventions tend to outperform policy tweaks on their own.

Example: AI-driven virtual try-on tackling fit-driven returns

An AI-driven virtual try-on approach illustrates one version of a fit-first stack. Shoppers upload a single front-facing photo, and the system generates a photorealistic image of how the garment fits their body within seconds, alongside a size recommendation derived from eight body measurements. That puts fit information directly at the point of purchase, rather than leaving shoppers to guess from a flat size chart.

Brands typically layer this alongside, not instead of, PDP work:

• Body-measurement sizing replaces guesswork based on past purchase history alone.
• Photorealistic try-on gives shoppers a visual answer to “will this fit me” before checkout, not after delivery.
• Installation as an app avoids a custom engineering project, and the same data can feed Klaviyo flows or returns analytics.

Any brand testing this should track its own before-and-after fit-return rate rather than relying on category averages alone.

Measuring success: KPIs, timelines and a simple ROI check

Four numbers matter more than any dashboard vanity metric: return rate by units and by revenue, the exchange-versus-refund split, cost per returned unit, and SKU-level return rate by size. Track them together, because a falling unit return rate alongside a rising cost-per-unit means you’re solving the easy returns and leaving the expensive ones.

Metric Why it matters Return rate (units/revenue) Headline health check, but revenue-weighted catches high-value SKU problems units alone miss Exchange vs refund split Rising exchanges with flat refunds signals bracketing is under control Cost per returned unit Processing a returned apparel item commonly runs into the tens of pounds once handling, restocking and markdown risk are counted SKU-level return rate by size Isolates chart errors from broader fit problems

Give any fit intervention at least four to six weeks and enough order volume per size to be confident the swing isn’t noise, particularly for lower-volume SKUs. A simple ROI check: (returns avoided × cost per return) minus (tool or programme cost) over the same period. Even a modest reduction in per-item processing cost compounds quickly across a full catalogue.

Pro Tip: Run your ROI calculation at the category level, not the storewide average — a size recommender that barely moves accessories returns can still pay for itself several times over on dresses alone.

Seasonal or trend-driven return reasons

Return patterns shift with the calendar, and treating every month the same hides real signal. Holiday gifting periods drive a spike in “didn’t fit the recipient” returns, since the buyer never had the chance to check sizing against the actual wearer. Swimwear and occasion-wear seasons bring higher bracketing rates, as shoppers order several sizes or styles for a single event with every intention of returning most of them.

Fast-moving trend items carry their own risk profile. A style that goes viral often sells to a broader, less brand-familiar audience than the core customer base, and that unfamiliarity with how a brand’s sizing runs shows up directly in the return rate. New-in ranges launched without full size-run data available yet tend to return higher than established bestsellers, simply because the size chart hasn’t been battle-tested by real order volume.

The practical response is to flag known seasonal spikes ahead of time and pre-brief customer service and returns operations rather than reacting after the fact. Gifting periods are also a good moment to promote exchange-first flows more heavily in return communications, since a mis-sized gift is one of the easiest returns to convert into an exchange rather than a lost sale. Trend and viral items deserve tighter early monitoring of SKU×size return data, precisely because the size chart hasn’t yet proven itself against a representative customer base.

Return rates variation by apparel category

Not all apparel returns at the same rate, and lumping tops, bottoms, and accessories into one blended number hides where the real problem sits. Dresses and fitted tops tend to sit at the higher end of return rates, since they carry more fit variables (bust, waist, length, sleeve) than a simple knit top. Bottoms, particularly trousers and jeans, run high too, because waist, inseam, and rise combine into more possible mismatch points than almost any other category.

Accessories, footwear aside, generally return at noticeably lower rates. A scarf, a belt, or a bag carries far fewer fit variables, so most of its returns trace back to product mismatch or damage rather than sizing. Footwear sits closer to apparel’s higher rates, since width and true-to-size variation between brands catches shoppers out constantly.

This variation matters for prioritisation. A brand spreading size-tech investment evenly across a catalogue is wasting effort on categories that were never the problem. SKU-mix data from broader industry returns reporting consistently shows dresses, swimwear, and fitted tops carrying disproportionate return weight, which is exactly where fit interventions should land first. Run your own SKU×category breakdown before assuming your catalogue mirrors the industry pattern.

Differences in return reasons between online and in-store purchases

Online and in-store returns aren’t the same problem wearing different clothes. Online apparel return rates typically sit around 23% to 25%, against roughly 9% in-store, and that gap comes down almost entirely to one thing: the ability to try something on before paying for it.

In-store returns tend to be due to genuine faults, damage discovered after purchase, or simple change of mind, with smaller share from fit issues since shoppers try items on before purchase. Online purchases face more fit uncertainty as the try-before-buy step occurs post-sale, with sizing decisions made from charts and photos instead of trying on.

This is why bracketing is almost entirely an online phenomenon. A shopper in a fitting room settles on one size and buys it. A shopper online, uncertain which of two sizes will fit, often orders both with a return in mind from the outset. It also explains why online return rates respond so strongly to fit technology and measurement tools, while in-store return reduction leans more on quality control and staff training. The two channels need genuinely different playbooks, not a single blended returns policy applied uniformly across both.

Influence of size guides and fitting technology on return reasons

A static size chart answers one question badly: it tells a shopper what a garment’s measurements are, but not whether those measurements suit their specific body. That gap is why traditional size guides, on their own, only make a modest dent in fit-related returns. They help the shoppers who already know their own measurements and are willing to compare them manually. Most shoppers do neither.

Fitting technology closes that gap by doing the comparison for the shopper. Size recommenders that ask for a handful of measurements or reference garments narrow the guesswork considerably, and virtual try-on tools that generate a visual preview go a step further by answering the fit question visually rather than numerically. The shift matters because shoppers don’t abandon a purchase over uncertainty as often when they can see a plausible answer in front of them, rather than interpreting a chart.

The honest caveat: fitting technology works best stacked on top of accurate base data, not instead of it. A size recommender fed inconsistent supplier measurements will confidently produce wrong answers. Brands that get the most out of this layer typically standardise their own measurement process first, using body-measurement data rather than purchase history alone , then layer a recommender or try-on tool on top of clean numbers.

Author perspective: sequencing the levers without hurting conversion

Fit-first is the correct order of operations, not because policy and QC don’t matter, but because they can’t outrun a chart problem. Fix the SKUs your data flags before touching global policy. Use SKU×size diagnostics to decide, case by case, whether a size chart, a photo set, or a returns policy is the actual lever, then test one change at a time and measure it honestly. Anything else is guessing with better spreadsheets.

— Jack

Where GarmCheck fits into a returns-reduction stack

Fit tech is one lever among several, but it’s the one that attacks the largest bucket in the ranking above. Some tools give merchants a way to put a size recommendation and a photorealistic try-on preview directly on the product page, built from eight body measurements rather than a generic chart or purchase history guesswork.

It stacks cleanly with the PDP and policy work covered earlier rather than replacing it: better photography and measurement tables still matter, and exchange-first flows still curb bracketing. What Garmcheck adds is the piece a static chart can’t provide, a direct answer to “will this fit me” at the moment a shopper is deciding whether to buy. Installation runs as an app, so there’s no engineering project to greenlight. If fit is showing up as your largest return category, start with the virtual try-on product page or check the size recommendation tool to see how the measurement flow works for your catalogue.

Sources

• The True Cost of Apparel Returns: Alarming Return Rates Require Loss‑Minimization Solutions
• Online clothing leads e-commerce returns with sizing driving most send‑backs (Retail Insider)
• Academic study on fashion returns drivers and prevention measures (Springer, 2021)

Recommended

• Reduce apparel returns: tactics that protect conversion
• Reduce Clothing Returns with AI Try-On
• Why 72% of Fashion Returns Are Fit Problems — And What to Do About It
• Virtual Try-On vs Size Guides: Why Size Guides Don’t Work — And What Does