Why are size charts so inaccurate across brands?
Size charts are inaccurate because there is no universal clothing size standard, and decades of vanity sizing have made the problem worse. Every brand defines its own sizing based on different fit models, target demographics, and design choices. A size 10 at Zara is not the same as a size 10 at H&M or ASOS — and the gap keeps widening.
A size chart only tells you what one brand thinks a "medium" is. It says nothing about whether that medium will fit your body, because the next brand's medium might be 5cm wider in the bust and 3cm shorter in the torso. Shoppers are left comparing numbers that don't mean the same thing across stores.
But inconsistency is only part of it. Size charts also flatten a multi-dimensional question ("will this garment work on my body?") into a single variable: a letter or number. That's where most fit-related returns start.
Vanity sizing: how much do clothing sizes vary between brands?
A "medium" can vary by up to 12cm in bust measurement across major fashion brands. That's not a rounding error. That's two or three sizes.
A ZAVA analysis of sizing across popular brands found that Mango's total size range spans 25.1 inches with roughly 2.5-inch steps between sizes, while Zara, All Saints, and H&M sit on the more restricted side with narrower ranges. In a survey of online shoppers, 65% of women reported finding big size differences between brands. Among men, 51% noticed variations.
Vanity sizing makes it worse. Brands intentionally label garments with smaller size numbers to flatter shoppers, and the drift has been going on for decades. A Time analysis of vintage sizing charts showed that size inflation accelerated through the 1980s, with some brands shifting the equivalent of 2-3 full size categories. A dress labeled "size 8" today would have been a "size 12" or "14" forty years ago.
So a shopper who's a size 10 in one store walks into another and finds the same label fits completely differently. The size chart was "accurate" in both cases. It just didn't mean the same thing.
What do size charts miss about garment fit?
Size charts ignore garment construction, and construction is what actually determines whether something fits well.
Two dresses both labeled "medium" can fit completely differently depending on their cut, fabric, and design. A fitted sheath dress needs a close body match in every dimension: bust, waist, hips, and length. A relaxed A-line is forgiving in most of those areas. A size chart treats them identically. Pick "medium" for both, and you'll have two very different experiences.
Design details matter more than overall size. Whether a top has set-in sleeves or dropped shoulders. Whether a waistline sits at the natural waist or higher. Whether the fabric has stretch. None of this appears in a standard size chart, and all of it changes whether a garment works on a given body.
Then there's personal preference. You might technically "fit" in a garment but find the waist tighter than you like or the neckline lower than you're comfortable with. Fit isn't purely dimensional. It's also about how you want clothes to feel.
The cost of inaccurate sizing: returns, lost revenue, and ecommerce churn
Inaccurate sizing costs the fashion industry an estimated $1 trillion per year in returns, according to RetailWire. Fashion ecommerce return rates average 30-40%, and fit issues drive over half of them. For dresses, return rates reach 50-55%.
A PRIME AI study analyzing 1 million fashion returns found that 67% were driven by fit and sizing issues. Here's what stood out: 71% of those customers said they checked the size chart before purchasing. The chart didn't prevent the return. In many cases it steered them to the wrong choice.
Each return costs retailers 15-30 euros in direct processing: return shipping, inspection, repackaging, restocking. But returned items often can't be resold at full price. Customer service time adds up. And shoppers who go through a return are 20-30% less likely to buy again within 90 days.
Then there's the cost nobody can measure: shoppers who never bought at all. They landed on the product page, weren't sure about the fit, and closed the tab. No return to process, no data to track, just revenue that quietly disappeared.
Do better size charts actually reduce returns?
Somewhat. Detailed sizing guides that include actual garment measurements (chest width, garment length, shoulder seam to seam) perform better than generic S/M/L tables. Adding fit descriptions like "relaxed through the body, fitted at the waist" gives shoppers more to go on.
But even good size charts hit two walls.
First, they need accurate self-measurement. A 1-2cm error with a soft tape measure at home can shift a shopper up or down a full garment size. Research published in BMC Public Health found that women tend to underestimate waist circumference, especially those with central fat distribution. Self-measured body dimensions can be off by 5-10cm from professional measurements. That's a big ask for a tool that depends on precise inputs.
Second, size charts can't express how a garment is designed to sit on the body. Whether it's meant to skim or cling, how stretch fabric behaves versus woven, how a dropped shoulder creates a different silhouette than a set-in sleeve. These details determine whether someone keeps or returns a garment, and a table of centimeters can't capture any of them.
What's replacing the size chart: virtual try-on, AI sizing, and fit finders
Several approaches try to match the specific garment to the specific shopper, rather than mapping measurements to a generic size label. They have different trade-offs.
| Approach | How it works | Reduces returns | Adoption friction |
|---|---|---|---|
| Model info on product pages | Model's height, weight, measurements, and size worn | Moderate (10-20%) | None (zero effort from shopper) |
| Detailed size guides | Actual garment measurements, fit descriptions | Modest (5-10%) | Low |
| Customer fit reviews | Reviewers share usual size + how item fit | Moderate (10-15%) | Low (passive) |
| Measurement-based AI | Shopper inputs body measurements, AI predicts size | Medium-high (15-25%) | High (tape measure) |
| Body scanning / photo | Phone camera captures body dimensions | Medium-high (15-25%) | High (photo upload) |
| Question-based fit finder / AI sizing | 3-6 simple questions about body and preferences | High (20-30%) | Low (15 seconds) |
One of the biggest recent shifts is model reference information. Stores like ASOS, Zara, and many DTC brands now show "Model is 175cm, 60kg, wearing size S" on every product page. This is enormously helpful because shoppers can compare themselves to someone whose dimensions they can actually see in the photos. It requires zero effort from the shopper and gives them a real reference point instead of an abstract centimeter table. Plenty of stores have found they can lean on this harder than on size charts.
A UK-based lingerie retailer that implemented AI-powered size prediction saw returns drop 27% and conversions rise 18%. Fashion brands using virtual try-on report fit-related returns falling 35-50%.
Friction is what separates tools that work in demos from tools that work at scale. Anything requiring body measurements or photo uploads gets low adoption because shoppers won't do it. Tools that ask a handful of simple questions ("are your shoulders broader, average, or narrower?", "does a fitted waist work for you?") get much higher engagement, which is the only way they actually move return rates.
The strongest implementations analyze each garment's construction (neckline, waist definition, sleeve type, fabric stretch) and match it against what the shopper already knows about their body. Instead of "you're a medium," the answer is "this dress has a fitted waist and V-neckline, and based on your proportions, it's a match." Or it's not a match, and here are three alternatives that are.
What fashion retailers should do to reduce size-related returns
Size charts aren't going away. Shoppers expect to see them, and they're still useful as a rough reference. But treating the size chart as the primary fit tool is leaving money on the table.
What helps:
- Show model dimensions on every product page: height, weight, and the size they're wearing. This is the lowest-effort, highest-impact change most stores can make today
- Add garment-specific measurements to size charts: chest width, garment length, shoulder seam to seam, not just generic S/M/L ranges
- Put fit descriptions on product pages: "relaxed fit through the body" or "runs narrow in the shoulders"
- Prompt for fit context in reviews: ask customers for their usual size and how the item fit compared to expectations
- Add AI-powered sizing: question-based tools that match the garment to the shopper take 15 seconds and cut returns 20-30%
The stores with the lowest return rates aren't winning because their size charts are better. They've added tools that give shoppers a real answer about fit before checkout, and the size chart is there if someone wants to double-check.
Frequently asked questions
- What is vanity sizing?
- Vanity sizing is the practice of labeling garments with smaller size numbers than the actual measurements warrant, so shoppers feel better about what they're buying. It's been going on for decades. A Time analysis of vintage size charts found that a dress labeled "size 8" today would have been a "size 12" or "14" forty years ago. Vanity sizing is one of the main reasons size charts fail across brands — the same label means different things at different stores, and the gap is structural rather than accidental.
- Why are clothing sizes different between brands?
- There is no universal sizing standard. Each brand develops sizing around its own fit models and target customers. A brand targeting 20-year-olds uses a different body template than one targeting 50-year-olds. On top of that, manufacturing tolerances mean it's impossible to produce every garment with identical dimensions, and vanity sizing pushes labels further apart.
- Why do I wear different sizes in different stores?
- Because brands define sizes differently. Vanity sizing varies by brand and price point, with higher-priced women's brands tending to run larger for the same label size. A size 10 at one store might be a 12 or 14 at another.
- Are size charts reliable for online shopping?
- They're a rough guide, not a reliable one. PRIME AI found that 71% of customers who returned items for fit reasons had checked the size chart before buying. Size charts can't account for how a garment is constructed, what your fit preferences are, or how far off your self-measurements might be.
- How can I find the right size when shopping online?
- Look for stores that go beyond the size chart. Question-based sizing tools that match the garment to your body are the most accurate option for most people. Customer reviews that mention the reviewer's usual size and how the item ran are also worth more than any centimeter table. If the store only gives you an S/M/L grid and a model's height, you're mostly guessing.