How To Tell If Shoes Are Fake: 8 Measured Checks

Fake shoes fail on measurements. A genuine size tag prints four rows, US sizes step 8.47 mm per size, and one style code repeats in three places.

How Fake Shoes Reach You

Anyone working out how to tell if shoes are fake is up against an industry, not a backroom. Counterfeit footwear travels in small parcels now, not containers. The OECD and EUIPO valued global trade in fake goods at USD 467 billion in 2021. Clothing, footwear and leather goods made up 62 percent of everything customs seized that year, and China accounted for 45 percent of reported seizures.

Three consequences follow from that scale, and they set the order of the checks on this page.

1. Logos fall first. A swoosh, a three stripe overlay or a silhouette costs a copy factory almost nothing to match. Judging a shoe by its outline is judging it on the one thing the copy gets right.
2. Codes fall second. A style code is a string of characters. Copy factories read real codes off real shoes and print them, so a valid code proves the code exists, not that your shoes do.
3. Measurements fall last. Lasts, insole lengths and stitch density cost money and tooling to match across a size run. This is where a copy line runs out of budget, and it is where a ruler beats an eye.

Two of the eight checks above need nothing but eyes and a phone. Four more need a ruler or callipers you already own. None of them need a photograph of a genuine pair for comparison, which matters, because a comparison photo of unknown provenance is worth nothing.

What Changes Between Brands and What Never Does

The shape of the product code changes with the brand. The rule that the code has to repeat does not change at all.

Nike is the strictest case and the easiest to test, because the six plus three structure breaks visibly when a copy gets the hyphen wrong. The Nike tag breakdown walks through the rest of that label, including the factory code and the barcode check digit. New Balance runs no separator at all and shifts between seven and eight characters by model, so agreement between the tongue label and the box label matters more than the count itself. The New Balance guide records the codes for the 990v6, the 2002R, the 9060 and the 550.

What never changes is the repetition. Every brand in the table above prints the code on the shoe and on the box, and publishes it on its own site. A seller who can photograph the shoe can photograph the box label. Ask for it. The absence of that photograph is a data point on its own, and it costs nothing to collect.

Model specific work starts once the code survives. The Jordan 1 checks count eyelet pairs and lace lengths, and the Air Force 1 checks measure heel stack and outsole thickness. Those numbers only exist because the model is fixed. On an unknown pair, the generic checks on this page are what you have. Everything in the sneakers section sits one level further down.

Size Tag Maths: The Step Between Sizes

Internal length moves by a fixed amount per size, and a factory that buys one last for a whole size run cannot hide it.

The US scale runs on the barleycorn, an old English unit of one third of an inch. Wikipedia’s shoe size entry records the barleycorn at 8.47 mm and the Paris point at 6.67 mm. Those two figures explain why a US size and a EU size never line up cleanly. The centimetre column is different again. Mondopoint describes the foot rather than the last, and brand charts round it to the half centimetre. That is why the Nike men’s footwear chart steps the centimetre row by exactly 5 mm per half size while the US step is 4.23 mm.

That rounding is the part most guides get wrong. The centimetre row is a pairing, not a measurement. Use it to test whether US 9 really sits against CM 27 on the label. Do not use it to check an insole with callipers, because the chart and the last disagree by design.

The insole test is the one worth the effort. Two pairs of the same model, one size apart, should differ by about 8 mm heel to toe. Copy lines commonly tool three or four lasts and cover eight or nine sizes with them, stretching the upper to take up the slack. The result measures identical across neighbouring sizes, and no amount of logo accuracy repairs it.

What These Checks Cannot Tell You

No measurement on this page proves a pair is genuine. Each one proves a specific copy failed.

A shoe that passes all eight checks has cleared the failures that cheap production lines make. The top tier of copies clears several of them too. Those lines buy the correct number of lasts, print the code they read off a retail pair, and run their stitching at retail density. Against that grade of copy, the useful evidence is provenance rather than geometry. Ask where the pair came from, what the seller can show of the purchase, and whether the box label barcode matches the size in the box.

Two honest limits are worth stating plainly. The stitch density band on this page rests on one trade source, which is why it carries a Low confidence rating rather than a High one. And the price floor check tells you whether an inspection happened, not what the inspector found.

Where To Get Your Shoes Authenticated

Four routes exist, and they differ in what they actually promise.

Marketplace verification is the cheapest, because the cost is already in the price. eBay’s Authenticity Guarantee for sneakers applies at 75 dollars and above, inspects within two business days of receipt, and attaches a tag with a QR code that opens the report. GOAT’s verification page describes a mix of digital authentication, in hand verification and machine learning built on what GOAT calls hundreds of thousands of logged data points. That page also states openly that some items pass without a physical inspection. StockX runs its own version, and the StockX review covers where that process is strong and where it is not.

Brand channels come second. Nike publishes a help article for buyers who think they bought a fake, titled Have I Bought Fake Nikes, and adidas publishes one titled How to notify adidas about fake products. Those pages put the brand on record about a seller. Neither is a same day verdict on the pair in your hallway, and neither replaces the marketplace route.

Photo pre screening sits before both. Legitly reads a pair from photographs and flags code, label and stitching problems in minutes, which is useful when the choice is whether to buy at all. It is a filter, not a certificate, and a pair that clears it can still fail an in hand inspection.

Independent human authenticators are the fourth option for a grail pair. Expect a fee per item and a written opinion rather than a guarantee. For anything under about 200 dollars, the marketplace route costs less and settles the question faster.

Verdict: When To Walk Away From a Pair

Walk when two or more measured values land outside their bands.

One value outside its range is a measurement problem as often as a counterfeit. Callipers slip, seams curve, and a worn insole reads short. Two values outside their ranges, on two different parts of the shoe, is a pattern. A size label with three rows and an insole that matches the size below it is not a pair of coincidences.

Three situations end the conversation whatever the other numbers say. The seller cannot produce the box label. The style code returns a different colourway. The listing sits a dollar or two below a marketplace authentication floor. Each of those is a deliberate choice by the person selling, not an accident of manufacturing.

Buy with the checks in mind rather than instead of them. Measure the pair when it lands, inside the return window, and keep the packaging until the numbers agree. A return costs postage. The alternative costs the whole purchase.