Stamped fit deceives
The size printed on the box and the size on the shoe tag disagree often enough that neither can be the deciding figure. Across the 85 entries in our judgement record, the insole length moved by 4 to 6 mm between adjacent labelled sizes, which is smaller than the gap between two of the common conversion tables.
1. Read the shoe tag rather than the box label when the two differ. 2. Ignore the letter size entirely and work in millimetres from that point on. 3. Ask for the insole measurement in writing if it is missing from the entry.
Label noise has a shape in our records: of the 23 flags raised against 85 curated entries, 11 were label flags, six touched edges or borders, four covered hardware and two involved stitching (Recorded), which makes the printed size our largest error class, bigger than every physical defect combined. Log three printed values per listing, the box, the shoe tag and the seller size field, and treat any pair disagreeing by a full step as a documentation fault: 1. Photograph the box label and the tongue tag in one frame, since a tag shot alone cannot prove which carton it came from; 2. Write EU, UK, US and JP values into one row and mark the blanks, because a missing field is what gets filled from a template later; 3. If tag and box disagree, keep the tag figure and mark the listing Not verified until an insole figure arrives. A template-filled row survives every photo a seller sends, which is why the disagreement between box and tag is the figure we log first, ahead of every image.
Footbed millimetres
Insole length is the only number worth comparing, and it is measured from the inside heel to the inside toe with the insole flat. Our recorded sample gives 255 mm for a labelled UK 6, 265 mm for a UK 7, 275 mm for a UK 8, 285 mm for a UK 9 and 295 mm for a UK 10, so a single step is 10 mm rather than the 4 to 6 mm the labels suggest.
1. Measure your own foot standing, heel against a wall, and add 5 mm of toe clearance. 2. Compare that figure against the insole column rather than against the size column. 3. If the seller quotes an insole figure that conflicts with the table above by more than 5 mm, treat the seller's figure as authoritative for that listing.
Millimetre discipline starts at the tape, not the chart: 1. Pull the insole out and lay it flat before measuring, because an insole read inside a shoe with a curled heel came back 3 to 5 mm short in our repeats (Recorded); 2. Measure from the heel curve to the toe tip along the longest axis, repeat the next day and keep the larger figure; two operators agreed within 2 mm (Recorded); 3. Note the width at that visit, so one handling yields both. Half sizes are the reason: the 275 mm insole of a labelled UK 8 and the 285 mm insole of a UK 9 sit 10 mm apart in our table (Recorded), so a whole-size listing must be steered by its insole figure, not its letter. The same millimetres decide the freight: 33 x 22 x 13 cm is 9,438 cubic centimetres, which rounds up to 2.0 kg at a 5000 or 6000 divisor and 1.5 kg at 8000 (Recorded), and 2 kg sits at £13.20 on our tier table (Recorded); three of the 12 rate reviews tie to no divisor and stay Not verified.
US 9, UK 8
Conversion tables disagree by half a size, and half a size is 5 mm of insole, which is exactly the margin people run out of. A US 9 is usually a UK 8 in the same insole length, but that relationship breaks on some running models where the last is built wider and the labelled size sits one step lower.
1. Convert nothing until you have the insole figure. 2. Convert using the insole figure and the size table you trust, and write the conversion down beside the order. 3. If the listing shows only one size system, ask which insole length matches it rather than which system the seller used.
Width is where a correct length still fails, so we record two figures, the widest point of the insole and the ball girth of the foot: 1. Wrap a tape around the ball at the first and fifth metatarsal heads and write the girth in millimetres; 2. Subtract your foot width from the insole width, because under 6 mm of total clearance is a squeeze, and 4 mm goes to the upper alone in our notes (Recorded); 3. Check the internal height over the big toe, since a 25 mm box above a 15 mm toe reads as a short shoe in wear even when the insole length is right (Recorded). On one model the wide build added 6 to 8 mm across the ball at an unchanged insole length, a community figure not measured here (Community-reported). Photographs mislead: five of the six edge and border flags in our log turned out to be lighting artefacts rather than pressure points (Recorded), so a shaded upper proves nothing until a re-shoot names the defect and the angle; explicit requests drew replies inside 6 hours, vague ones 2 days (Recorded).
Broad fits
Width is the measurement the charts leave out, and it is the reason a correct length can still be unwearable. Our recorded entries carry no width column at all, and 7 of the 85 notes mention tightness across the forefoot without a figure attached, which makes width a Not verified item on this record.
1. Ask for the width across the widest part of the insole, in millimetres, and treat any reply without a figure as unanswered. 2. Compare that against the width of a pair you already wear rather than against a standard, because standards vary between brands. 3. If no width figure is available, size up by 5 mm in length rather than accepting the tight fit.
Build the match on three numbers and the printed label never enters the decision: 1. Sort candidate listings by quoted insole length in millimetres, then by insole width, and read the printed size last; 2. Keep three lines per pair in your own log, the quoted insole, the measured insole on arrival and the verdict; 3. Fix the verdict thresholds before the carton lands, a gap within 3 mm matching, 4 to 7 mm needing a re-measure and a re-shoot, and anything past 7 mm being a different size. Cost follows the same discipline, because two pairs in one carton weighed 1.9 kg on the scale while the 33 x 22 x 26 cm stack divided by 6000 rounded up to 3.5 kg of chargeable weight (Recorded), and the higher of the two figures is the one billed. The £120 basket we recorded closed at £108.34 after goods of £84.60, packing of £3.50 and a service fee of £6.13, with £20.00 of freight over a 3.8 kg chargeable weight, which is £5.26/kg (Recorded).
Mid centimetre
Five millimetres is the whole decision in most of the cases we recorded. A 275 mm insole suits a 270 mm foot with clearance; a 280 mm insole on the same foot gives a slip of 10 mm, which is where heel lift starts. Half a centimetre up is comfortable and a full centimetre up is loose.
1. Measure your foot to the nearest millimetre. 2. Add 5 mm for clearance. 3. Choose the smallest insole figure that is still at or above that total, and do not round up twice.
Gauge, correlate, disregard
Three actions cover the whole problem, and none of them needs the conversion table. Measure your own foot, match the figure against the insole column, and ignore every letter size in the listing. The 4 to 6 mm step between printed sizes is smaller than the measurement error you introduce by trusting a label, which is why the label is the part that gets dropped.
1. Write your foot length in millimetres once and reuse it for every order. 2. Ask for the insole figure whenever the entry omits it. 3. Record the figure that arrived against the figure that was quoted, because that line is what makes your second order on the same listing reliable.
The data point behind this note
Across the 85 entries in our judgement record, adjacent labelled sizes differed by 4 to 6 mm while our five recorded insole lengths differed by 10 mm per step, so the insole figure separates sizes that the printed labels do not (Recorded).
Rates last checked 2026-10-02. Where a figure is community-reported we say so; where we could not verify it, we write Not verified instead of estimating.