There is a private ritual most women have performed at some point in a fitting room, usually under the slightly green-tinted lighting that retail still inexplicably uses. You bring three pieces in, all labelled the same size, all from the same general category of brand. One is loose at the waist. One does not button. One looks, somehow, completely different on your body than the model in the lookbook. You stand there, half-undressed, and you do the thing women have been quietly trained to do for fifty years.
You blame yourself.
Maybe the second one is closer to your real size. Maybe you have gained weight without realising. Maybe the first one is misleading you. Maybe you should not have eaten lunch. Maybe your body is the problem.
The body is almost never the problem. The clothes are. And the entire system of women's clothing sizes in 2026 is built on a quiet, multi-decade lie that the fashion industry has every commercial reason to keep telling and very little reason to fix.
This is the piece nobody wants to write, because it implicates almost every retailer simultaneously. But once you understand what has actually happened to clothing sizes since the 1940s, the experience of standing in a fitting room with three pieces that all fit differently stops being a private failure. It becomes what it actually is. A documented industry-wide failure of measurement.
The number that should make you angry
Start with the single most quoted statistic in the history of vanity sizing, because it is the clearest possible illustration of what has happened. A women's size 14 in 1937 was equivalent, by body measurement, to a women's size 8 in 1967. That same body, measured against today's typical American retailer's size chart, would wear a size 0 — or, in many brands, a 00.
Read that sentence again. The same woman, with the same body, wearing the same dimensions, has gone from a size 14 to a size 0 in seventy-five years — without losing a pound. Her body did not change. The labels on the clothes did. And the labels did not slowly drift smaller because women's bodies were getting smaller. They drifted smaller because, somewhere along the way, the fashion industry discovered that women bought more clothes when they thought they were wearing a smaller number.
This is documented. In the fifty years between 1958 and 2008, a US size 8 expanded by up to six inches in actual measurement. Between 1937 and the present, the size labels have moved by something like six to eight full size categories. The shift is so well-known inside the industry that there is even a term for it. It is called size creep, and you can find it discussed openly in academic papers, in pattern-making textbooks, and on the technical pages of fit consultancies that sell their services to the brands doing the creeping.
None of this is on the labels of the clothes you buy.
How sizing actually worked, before it stopped working
It is worth understanding how the system was supposed to function, because the breakdown is what explains the present chaos.
In the late 1930s and early 1940s, the United States Works Projects Administration funded a major study of American women's body measurements, aiming to create a standardised national sizing system for ready-to-wear clothing. The intention was practical — the mass-production of clothing required predictable measurements, and women were tired of buying clothes that did not fit. The study produced the first National Bureau of Standards women's sizing guidelines, published in 1958.
That system had serious flaws even at launch. It was built on measurements of mostly white, young, slim women — the sample was biased toward what researchers could conveniently measure, not toward what the American population actually looked like. The resulting size 14 was smaller than the actual American median. But it was at least a system. Brands followed it. Sizing was relatively predictable. A woman could walk into one store and walk into another and reasonably expect that her size would mean something close to the same thing in both places.
By 1983, that system was officially abandoned. The standards became voluntary. Brands were free to make their own size charts. The age of vanity sizing began, and forty-three years later, it has not ended. Every retailer now operates a private internal sizing system tied to a private internal fit model — typically a single human woman whose dimensions are graded up and down to produce the full size range. The fit model varies by brand. The grading scale varies by brand. The interpretation of what a size 6 actually means varies by brand. A size 6 at Reformation is not the same body as a size 6 at Madewell, which is not the same body as a size 6 at Zara, which is not the same body as a size 6 at any independent designer producing in Lisbon or Buenos Aires or Brooklyn.
This is the system every shopper is currently operating inside. It is, by any reasonable definition, not a system at all.
Why vanity sizing keeps getting worse, not better
The natural assumption is that as ecommerce matured and returns became expensive, brands would have a strong incentive to fix this. Returns cost the apparel industry approximately forty-five billion dollars a year, and fifty-two percent of apparel returns are size-related. The financial logic should be obvious. Standardise sizes, reduce returns, save billions.
It has not happened, for three reasons that are worth naming.
The first is brand differentiation. Vanity sizing functions as a competitive tool. The brand that labels a body as a smaller number than its competitor wins the emotional moment in the fitting room. Customers do not buy from brands that make them feel large. They buy from brands that make them feel slim. This is so well understood inside the industry that some brands deliberately label their sizes more generously than the competition as part of their positioning. The retail tactic depends on the deception continuing.
The second is fit-model dependency. Most brands build their patterns around a single fit model whose body becomes the template. If that model is five-foot-nine, slim through the torso, narrow at the hip, the entire size range will reflect that proportion — graded up and down mathematically, but without re-measuring real bodies at each size. The result is that the patterns work for people who happen to share the fit model's proportions and fail for everyone else. A woman whose body is shaped differently from the brand's fit model will find that no size in that brand fits her, regardless of how much weight she gains or loses.
The third is global supply chain fragmentation. Different countries have different sizing standards. European sizes, UK sizes, US sizes, Japanese sizes, Chinese sizes — each system uses different reference measurements, different grading scales, different labelling conventions. A piece of fast fashion that crosses three supply chains to reach you will have been sewn against a Chinese factory's interpretation of a Western brand's size chart, with measurements taken at different points on the body than your local retailer would use, and graded by different rules. The chaos compounds.
What this does to bodies and minds
The financial cost of all of this is brutal but quantifiable. Forty-five billion dollars in returns. Roughly thirty percent of all online apparel orders sent back to the warehouse. Mountains of fashion landfill created largely by the sizing chaos rather than by genuine quality problems with the underlying clothes.
The psychological cost is less visible and probably more serious.
Decades of inconsistent sizing have trained women, in particular, to interpret a clothing label as information about their bodies rather than information about the clothing. When a piece does not fit, the assumption defaults to a body explanation rather than a manufacturing explanation. The clothes are presumed correct. The body is presumed wrong. This is the inversion that makes vanity sizing structurally toxic. It is not just that the numbers are misleading. It is that the misleading numbers are then interpreted as accurate descriptions of the wearer.
The cumulative effect, across a lifetime of fitting rooms, is a quiet erosion of trust in your own body's reliability. You stop knowing what size you are because what size you are depends on which brand you happen to be standing in. Your relationship with your reflection becomes mediated by an arbitrary tag that has nothing to do with your actual dimensions. The fitting room becomes a place of small daily betrayals, all of them invented by an industry that profits from your confusion.
What actually fits, and how to find it
The path out of this is technical, slightly tedious, and once you implement it, transformative. It comes down to three steps.
One. Know your actual measurements. Not your dress size. Not your jean size. The actual numbers. Bust at the fullest point. Underbust. Waist at the smallest natural point. High hip. Full hip at the widest point. Inseam from crotch to floor. Shoulder to shoulder. Arm length. Take them with a soft tape measure, write them on a single card, keep them in your phone. These numbers are the only honest description of your body that exists. Everything else is interpretation.
Two. Shop the measurements, not the labels. Almost every reputable retailer now publishes garment measurements in addition to size labels — the actual centimetre or inch dimensions of the piece across the chest, waist, hip, and length. These numbers do not lie. The label might say size 8; the measurements will tell you whether the actual garment will fit your actual body. The size system as a whole is broken. The garment measurements are not. They are the workaround.
Three. Build a vocabulary for what works on you. Not what size you wear. What proportions you wear. A woman with a long torso and shorter legs will have a completely different ideal fit profile than a woman of the same height with the opposite proportions. The size label cannot capture this. But once you know your proportions — the actual length of your torso relative to your legs, the actual relationship between your bust, waist and hip measurements, the actual cut of trouser that works for your specific build — you stop being a customer of the sizing system and start being a customer of clothing that fits.
Where the system finally starts to bend
If there is any structural good news in this story, it is that the new infrastructure of AI shopping — the same agentic commerce shift we wrote about yesterday — has the potential to dismantle the worst of the sizing problem in the next few years.
The technology already exists. Tools like Bold Metrics use machine learning to predict more than fifty body measurements from a simple height and weight input, then match those measurements against garment-specific data from individual brands. Gap recently became the first major retailer to integrate this into an agentic commerce flow, allowing in-chat size recommendations that take the brand's specific fit model into account rather than the meaningless label number. An Agentic Sizing Protocol — an industry-wide API that lets AI agents query fit data in real time — is in active development.
What this means in practice is that the next generation of shopping, if it works as designed, will route around the broken size label entirely. You will tell an AI agent your measurements once. The agent will check those measurements against every brand's actual garment dimensions and return only the pieces that will actually fit. The size on the tag will become irrelevant. The body becomes the constant. The clothing adjusts.
Whether this future actually arrives, or whether the industry finds new ways to gatekeep the data so the deception continues, is the question of the next five years. But for the first time since 1983, the technology to fix the system exists.
The reframe nobody puts on the marketing
Here is the part to take home, because it changes what happens the next time you stand in a fitting room with three pieces in the same size that all fit differently.
Your body is not the variable. Your body is the constant. The clothes are the variable. They have been the variable the entire time. The fact that women have been trained to read the variable as the constant and the constant as the variable is one of the longest-running quiet manipulations in consumer culture, and it is not your job to keep falling for it.
The next time something does not fit, the question worth asking is not what is wrong with my body. It is what is wrong with this garment's measurement system. Sometimes the answer is the cut. Sometimes it is the brand's fit model. Sometimes it is a deliberately-flattering vanity-sized label that misled you into thinking the piece would work on someone with your actual dimensions. None of these are your problem to fix. They are problems the industry created and is finally, slowly, starting to address — mostly under commercial pressure rather than out of any sudden moral clarity.
The independent designer world, it is worth noting, has historically been better at this than the mass market. Small ateliers producing in batches of forty or fifty do not have the volume to abuse vanity sizing the way scale retailers do. Their measurements tend to be more honest, their fit models tend to vary more, and the smaller the brand, the more likely you are to find clothing that respects the body of the person buying it rather than gaming the label for emotional manipulation in the fitting room.
That is not a guarantee. It is a tendency. But it is one of several quiet reasons that the people who shop independent designers, secondhand, and craft-based brands often report a calmer, more accurate, less psychologically loaded relationship with sizing than the people who shop primarily at mass-market retailers. The label deception was never built into their model. Their clothes are allowed to fit the bodies that buy them.
You have not been a difficult customer. You have not been in the wrong size. You have been a thoughtful person trying to operate inside a measurement system that the industry openly admits is broken and has no incentive to fix on its own. Knowing that does not make the next fitting room visit easier. But it does, finally, put the blame in the right place.
Frequently Asked Questions
What is vanity sizing?
Vanity sizing is the practice of labelling garments with smaller size numbers than their actual measurements would historically suggest, in order to flatter customers and increase the likelihood of purchase. A women's size 14 in 1937 was equivalent to a size 8 in 1967 and, in many present-day American brands, a size 0. The body has not changed; the label has shifted dramatically over time.
Why are clothing sizes so inconsistent between brands?
Every brand operates a private internal sizing system tied to a private fit model — typically a single human woman whose body becomes the template for the entire size range. The fit model varies, the grading scale varies, and the interpretation of what each size number means varies. National sizing standards in the US were officially abandoned as binding in 1983 and have remained voluntary ever since.
How much does sizing chaos cost the fashion industry?
Fifty-two percent of apparel returns are size-related, accounting for approximately forty-five billion dollars in industry losses each year. Roughly thirty percent of all online apparel orders are returned, the majority for fit reasons. The financial incentive to fix the system is enormous, but commercial incentives to keep vanity sizing in place have so far outweighed the cost of returns.
What is the best way to shop in the current sizing environment?
Know your actual body measurements — bust, underbust, waist, high hip, full hip, inseam, shoulder, arm length — and shop against published garment measurements rather than size labels. Most reputable retailers publish actual centimetre or inch dimensions for each garment, which do not lie even when the size tag does. Build a personal vocabulary of which proportions and cuts work for your specific build rather than which size you wear.
Will AI shopping fix the sizing problem?
The technology to fix it exists. Tools like Bold Metrics already predict more than fifty body measurements from simple inputs and match them against brand-specific garment data. Major retailers including Gap are beginning to integrate this into AI shopping flows. Whether the industry adopts this at scale or finds new ways to preserve the existing deception is the open question of the next several years.
