Industry Insights
Sneaker Sizing and Grading: How One Sample Becomes a Full Size Run
How sneaker sizing and grading works—last grading, pattern grading, width rules, and the fit-testing that prevents size-run complaints.
# Sneaker Sizing and Grading: How One Sample Becomes a Full Size Run
A sneaker is approved in one size—then multiplied into a full size run by a process most brands never think about until the complaints arrive: "runs small," "size 11 crushes my toes," "the same size fits differently in two colorways." That process is grading, and its quality decides whether every customer in every size receives the shoe you approved or a distorted version of it. This guide explains how sizing and grading work, where they go wrong, and the testing that keeps a size run honest.
Two Gradings, One Fit
"Grading" actually happens twice, in two places that must agree:
Last grading. The shoe last is scaled across the size run. Each size's last is a scaled form—length proportional, but width, instep height, and volume adjusted by rules that reflect how real feet change with size. Feet don't scale linearly: a foot two sizes longer is not proportionally wider, and grading rules encode that non-linearity.
Pattern grading. The upper pattern is scaled to match each last size. Pattern grading must track the last's rules exactly, or the upper fights the last— producing size-dependent wrinkles, pressure points, or sloppiness.
When the two gradings disagree, the symptom is precise and diagnostic: the same labeled size fits differently across the run, or fit quality degrades toward the extremes. This is why our shoe last guide insists the last decision comes first—the last's grading rules cascade into everything.
The Grading Rules That Matter
Four rule families determine grading quality:
Rule family / What it controls / Failure symptom
• Length steps — Size-to-size length increments — Inconsistent size gaps; "between sizes" confusion
• Width increments — How width grows with length — Big sizes feel narrow; small sizes feel wide
• Volume adjustments — Instep height and toe box growth with size — Large sizes crush the instep or toes
• Proportion limits — How much silhouette character scales — The size 13 looks like a different shoe than the size 8
The classic errors live in width and volume: proportional scaling (the lazy default) makes large sizes proportionally too narrow and small sizes too wide—exactly matching the most common size-run complaints. Good grading uses width and volume rules that grow more slowly than length. Ask a factory how their grading handles width versus length; a fluent answer is a capability signal, consistent with the supplier verification habit of probing process knowledge.
Size Systems and Conversion
Small brands selling internationally meet the size-system maze: US men's and women's, UK, EU, and the Mondopoint (mm) system that factories actually work in. The practical rules:
• The factory's truth is millimeters. Lasts and patterns are dimensioned in mm (Mondopoint); all other systems are labels mapped onto that truth. Get the mm-to-label mapping from the factory in writing and publish it—this becomes your size chart's foundation.
• Build your own conversion chart. Generic internet conversion charts vary; your brand's chart should come from your last's actual measurements, which is also how you make the size chart page a trust asset rather than a copy-paste.
• Watch women's sizing. If the line serves women's sizing, that is typically a separate last or a carefully validated mapping—not a label swap. Fit complaints from mis-mapped women's sizes are among the most preventable in the industry.
Fit-Testing a Size Run
Grading is verified the same way everything fit-related is verified: on feet, at the extremes. The protocol that catches grading errors:
1. Fit three sizes minimum: smallest, largest, and the sample size. Grading errors concentrate at the extremes; the sample size was already approved.
2. Wear them, walk in them, measure complaints in dimensions. "Toe box tight by ~5 mm in size 13" is an actionable grading correction; "big sizes feel bad" is not.
3. Check the same size across colorways and materials. A stretchy knit upper and a stiff leather upper on the same last fit differently; material-driven fit variance is real and should be sampled per material family, not per colorway.
4. Log and correct at the grading-rules level, not by remaking one size. A one-size fix that ignores the rule producing it will resurface in the next size.
This protocol belongs in the sample rounds plan—not after bulk. Fit corrections discovered after production size-run are the most expensive kind.
The Size-Run Business Decision
Beyond fit mechanics, the size run is an inventory decision. Every size in every colorway is stock at risk; the run's shape should follow your market:
• Concentrated runs (fewer sizes, more depth per size) suit focused audiences and small budgets—fewer SKUs, less tail risk.
• Full runs serve broad channels and retail partners, accepting tail-size risk for coverage.
• The distribution curve matters more than the count: real sales cluster in the middle sizes. Overstocking the extreme sizes—the most common new-brand error—converts tail sizes into permanent inventory.
Our SKU architecture guidance covers the SKU-count discipline; the size run is that discipline's most granular layer. Combine the two: a tight silhouette count with a market-shaped size run beats a wide range with a naive one.
Grading and Size Runs with a Manufacturing Partner
Grading quality is invisible until worn, which makes it a verification-heavy process: checking the factory's grading rules, fit-testing the extreme sizes, and holding the mm-to-label mapping in writing. From the Fujian belt we run exactly that loop—grading rules reviewed with the factory's pattern team, extreme-size fit sessions arranged, size charts built from actual last measurements, and per-material fit variance checked before bulk. Our how to develop a sneaker guide shows where this sits in the full sequence.
On minimums: "MOQs depend on the product and factory. Tell us what you need and we confirm what is realistic before you commit to anything."
FAQ
Why do large sizes often feel narrower? Because proportional scaling—the lazy grading default—scales width as fast as length, while real feet widen more slowly than they lengthen. Good grading uses width rules that grow more gradually. The fix is at the grading-rules level, verified by fit-testing the largest size, not by remaking individual pairs.
What is Mondopoint sizing? The millimeter-based system factories actually work in: shoe size expressed as foot length in mm. US, UK, and EU sizes are labels mapped onto Mondopoint truth. Build your size chart from your last's actual millimeter measurements with your factory's mapping in writing—generic conversion charts vary and erode buyer trust.
How do I test that my size run fits correctly? Fit-test at least three sizes: smallest, largest, and the approved sample size—on real feet, walking, with complaints measured in dimensions. Also check the same size across material families, since stretchy and stiff uppers fit differently on the same last. Extremes are where grading errors concentrate.
Should every size be stocked in every colorway? Not necessarily. Real sales cluster in middle sizes; overstocking extreme sizes is the most common new-brand inventory error. Shape the size run to your market's distribution and your channels—concentrated runs suit focused audiences, full runs suit broad retail—with the SKU architecture discipline applied at the size level.
