A medium in one brand. A large in the next. A size 12 that fits like a 10 and a size 10 that fits like a tent. Online clothes shopping runs on a sizing system that is not a system at all — every brand grades its own ruler, vanity sizing drifts the numbers year by year, and the letter on the tag carries almost no information. The shoppers who order calmly and return rarely are not lucky. They have stopped making five specific mistakes, and the mistakes are worth naming first, because the fixes only make sense against them.
Mistake 1: Ordering the letter you usually wear
The letter is a rumor. An M at a Scandinavian brand, an M at an American mall brand and an M at a fast-fashion label can span four inches at the chest. Ordering “your size” online is ordering a guess with a logo on it. The fix is to ignore the letter entirely and read the garment measurements on the size chart — the actual inches for that specific item, not the generic brand chart, since better retailers publish per-garment numbers. Two garments both labeled M may differ by two inches at the waist; the chart tells you, the letter never will. This single habit — chart before letter, item chart before brand chart — removes most online sizing drama at the source.
Mistake 2: Never having measured yourself
The chart is useless without your own numbers, and most people have never taken them. Five minutes with a soft tape gives you the set that unlocks every size chart on the internet: chest or bust at the fullest point, natural waist, hips at the fullest point, inseam, shoulder width, and arm length if you buy tailored pieces. Measure over thin clothing, tape snug but not digging, and write the numbers in your phone where they will live permanently. Guidance on technique — where exactly the tape sits for each measurement — is set out clearly in MasterClass’s guide to measuring yourself for clothing, and it is worth doing once, properly. Re-measure once or twice a year; bodies change and charts do not care about last year’s numbers. Armed with real measurements, the between-sizes question also answers itself: you are not between sizes, you are a number, and the garment either contains it or does not. The same measurement habit pays off in the accessories aisle, where the method in how to pick the right belt size for your waist turns another guessing game into arithmetic.
Mistake 3: Skipping the reviews, which are a fit database
Reviews are the only place where a garment’s chart meets actual bodies. Read them for fit intelligence, not for stars: “runs small,” “size up if between,” “long in the torso,” “fabric has no give” are data. The most valuable reviewers are the ones who state their height, weight and the size they bought — a reviewer built like you, reporting that the medium fit perfectly, is worth more than the entire chart. Sort or search reviews for those reports, and discount the extremes at both ends, which are usually about shipping and moods rather than fit. If a listing has no reviews and no garment measurements, treat it as a gamble priced accordingly: fine for a $20 experiment, unwise for anything you would be sad to send back.
Mistake 4: Ignoring the fabric’s forgiveness
Two garments with identical charts fit differently because fabric is a variable the chart does not capture. A knit with 5% elastane forgives an inch in every direction; a rigid woven forgives nothing. Read the fiber content like a fit feature: stretch blends, smocked panels, elastic backs and adjustable straps all widen the target, while structured wovens, non-stretch denim and fitted linings shrink it. Your personal fit profile matters here too — where you carry weight, whether you prefer ease or closeness, whether your shoulders or hips are the limiting measurement. Note, per brand, what worked: a simple list in your phone (“Brand X trousers: 31, runs long; Brand Y dresses: 12, size up”) becomes, within a year, a personal sizing database no algorithm can match, because it is built on the only body involved.
Mistake 5: Gambling one size on an expensive piece
For investment pieces — the coat, the suit, the boots you intend to keep for years — order two sizes when the retailer offers free returns, and keep the better one. The arithmetic is blunt: a $0 return beats a $200 mistake, and the difference between adjacent sizes on a structured garment is often the difference between keeping it for a decade and resenting it by spring. Try both on at home, on carpet, with the shoes and layers you will actually wear, in daylight. Then return the loser immediately; a return that sits in a bag by the door for three weeks has a way of becoming an expensive permanent resident. Virtual fitting tools, the quizzes and avatar widgets now common on retail sites, help at the margins: they are reasonable at flagging a wildly wrong size and unreliable at fine distinctions, so treat their output as a second opinion behind your measurements, your brand notes and the reviewers who share your build. Between sizes, the standing rule: size for the largest measurement and tailor the rest, size up in rigid fabrics, size down only in stretch. The letter was never the answer. The numbers were.
Mistake 6: Treating shoes and one-size items as chart-free
The measurement discipline matters most where shoppers abandon it first. Shoe sizes convert between US, UK and EU systems with half-size disagreements that vary by brand, and a boot that will be worn with thick socks is a different size from the same boot in summer — measure your foot in centimeters, in the evening, standing, and order from the centimeter column where one exists. Bras, swimwear and anything sized by a second measurement (waist and length, bust and band) deserve the same double-number honesty. And “one size” garments are not size-free; they are size-concealed, usually cut for a narrow band in the middle of the range — read the model’s stated size and height on the listing, a data point retailers now commonly publish, and treat it as the closest thing to a chart the item will ever give you.
Mistake 7: Learning nothing from what goes back
A return is data wearing a shipping label, and most shoppers throw the data away with the box. Every return has a cause in a small taxonomy: chart lied, fabric had no give, cut wrong for your proportions, length misjudged, quality disappointed. Spend twenty seconds assigning the cause and noting it against the brand — the phone note from Mistake 4 grows a second column — and your future error rate falls in a way no fitting technology can match, because it is calibrated to the interaction between one body and one brand’s habits. The companion habit is keeping the winners’ receipts of information: when a garment fits perfectly, record brand, size, and the garment’s chart measurements at the moment of triumph, not from memory later. A year of this produces the personal sizing database mentioned earlier; two years of it makes online sizing feel rigged in your favor, because for you, uniquely, it is.
The special cases: shoes, bras and the genuinely hard categories
Two categories deserve their own footnotes because their sizing physics differ. Shoes sized by number hide a measurement the ritual can still recover: your foot’s length in centimeters, traced standing in the evening (feet swell through the day), converts through the brand’s chart more honestly than your usual number, and width — the letter beside the number that most shoppers ignore — decides comfort more than length does. For bras and other precision-fit garments, the honest advice is that the first purchase in a new brand belongs in a physical fitting room whenever one exists; get professionally fitted once, record the numbers, and only then does online buying become the measurement game the ritual can win. The ritual’s real claim was never that everything should be bought unseen. It is that what must be bought unseen can be bought by evidence — and that knowing which categories deserve a fitting room is itself part of the evidence.
The pre-order ritual, assembled
The fixes compress into a two-minute ritual that runs before any checkout. One: open the item’s chart, not the brand’s. Two: compare your three relevant numbers — from the note in your phone — against the garment’s, and choose the size whose numbers contain yours. Three: skim reviews sorted for fit comments from people built like you. Four: check the fiber content for forgiveness, and the returns policy for cost. Five: for anything over a set threshold — your own number, honestly chosen — order the second size if returns are free. Run the ritual and the outcomes change category: returns become the exception, kept pieces fit on the first wear, and the wardrobe stops accumulating almost-right garments that were almost returned.
