A fashion shoe forgives. If it is a few millimetres tight across the forefoot, the leather gives within a week. If the toe box is short, the shoe is worn for special occasions and quietly retired. A safety shoe does neither. It has a metal or composite cap over the toes that does not stretch, a plate in the sole that does not flex, and it is worn for ten hours a day, five days a week, on concrete. It is also, in most cases, chosen from a list by someone other than the person wearing it.
That combination — rigid geometry, long exposure, and a purchase decision made at a distance — makes safety footwear the most demanding fit problem in the industry. It is also the least discussed. The conversation around safety shoes is almost entirely about protection, because that is what the standard regulates. Fit is left to the shop floor.
What the standard actually covers
EN ISO 20345 is the European standard for safety footwear. It defines the protective classes — SB, S1, S1P, S2, S3 and, in the 2022 edition, S6 and S7 — and a long list of additional markings: penetration resistance, energy absorption in the heel, antistatic behaviour, water resistance, slip resistance, heat and cold insulation, cut resistance, ladder grip, scuff cap.
It is a protection standard. It says nothing about whether the shoe fits the person wearing it.
Read through the geometry of a foot, though, a handful of those markings turn out to change the shape of the fit problem. The rest are performance properties that leave the inside of the shoe untouched.
- The toe cap Present in every safety class. It must withstand a 200-joule impact and a 15-kilonewton compression, and the standard specifies a minimum internal clearance that must survive the impact test, per size. That clearance is the real front wall of the shoe. It is not the last's toe shape, and it is not the leather.
- Penetration resistance (P, PL, PS) A steel or textile insert across the sole. Steel in particular removes most of the sole's ability to flex at the ball of the foot. A foot that cannot roll through the step needs the length allocated differently than one that can.
- Metatarsal protection (M) A rigid or semi-rigid plate over the instep. Where it is present, the instep — normally the most forgiving zone in a laced shoe — becomes a hard boundary like the toe cap.
- Ankle protection and insulation (AN, CI, HI) Padding at the collar and thicker linings. Both reduce the internal volume of the shoe relative to the last it was built on.
Everything else in the standard — slip, antistatic, fuel resistance, heat-resistant outsole, water resistance — matters enormously for safety and not at all for fit.
Why the toe cap changes the problem
In a conventional shoe, the space in front of the toes is set by the last, and the last is a smooth, tapering form graded in fine steps from one size to the next. The upper around it has some give. If the foot is slightly long for the last, the shoe accommodates, and the wearer notices little.
A toe cap has none of these properties. It is a fixed cavity of steel, aluminium or composite. It does not deform under the foot. And because caps are manufactured components rather than shaped forms, they are typically graded in coarser steps than the last around them — one cap may serve two or three sizes.
The front wall of a safety shoe is not a shape. It is a boundary. Fit is defined by whether the foot reaches it.
This has a practical consequence for anyone trying to predict fit. For a fashion shoe, the question at the toes is how much room, and does the material help. For a safety shoe, the question is binary first — does the longest toe clear the cap — and only afterwards a matter of comfort. A model built for the first question will be confidently wrong about the second.
What the standard does not say
The parameters that most often decide whether a safety shoe is comfortable at the end of a shift are not in EN ISO 20345 at all. They are contextual, and they are usually invisible at the point of purchase.
- The sock Feet are measured bare. Work socks are thick — often two to three millimetres of loop-pile wool or technical fibre. Every width and girth measurement is effectively that much larger once the sock is on. Retailers' own fitting advice says to try shoes in the socks you wear at work; a fit model has to make the same allowance.
- The insole Safety shoes increasingly ship with a thick, shaped, removable insole, and wearers frequently replace it with an orthopaedic one. Each swap changes the height available over the instep by several millimetres, and in some jurisdictions an uncertified insole voids the shoe's certification. The shoe's inside is not a fixed quantity.
- The shift Foot volume rises measurably over a long working day on hard floors. A shoe that fits at eight in the morning is a different fit at five. Recommendations for safety footwear need to be made for the end of the day, not the start.
- The closure Dial-and-cable systems have become common in safety footwear. They apply even tension across the whole instep and adjust in fine increments — a closure that behaves like a well-tied lace, but more consistently. It is a different fit envelope from a traditional lace, and different again from a pull-on boot with no closure at all.
- The last's gender A large share of safety footwear is built on a men's last and sold in the full size range. A woman's foot in a men's last is systematically narrower at the heel and lower over the instep than the shoe expects. The shoe is not wrong; it was simply designed for someone else.
- Sole stiffness Beyond the penetration insert, the sole compound and construction determine how much the foot can roll through the step. Stiffer soles ask for more length. This is rarely stated on a product page and never in the standard.
Why it matters more here
In fashion retail, a poor fit produces a return. It is a cost, it is measurable, and it is the number the industry has organised around.
In safety footwear, a poor fit often produces nothing visible at all. The shoes are issued under a framework agreement, the size was picked from a chart, and the wearer keeps them because that is the pair they have been given. What follows is not a return. It is blisters in the first week, altered gait by the first month, and a foot, knee or back complaint that surfaces months later and is never traced back to a shoe that was one size out and two millimetres too narrow.
The people who buy safety footwear — employers, procurement, safety officers — are not measuring returns. They are measuring sick days, near misses and the cost of re-issuing equipment. Fit sits underneath all three, unrecorded.
A fashion shoe that fits badly comes back. A safety shoe that fits badly gets worn.
What we don't know yet
We want to be clear about where the ground is firm and where it is not.
The mechanics above are not in dispute. A toe cap is a boundary; a steel insert stiffens a sole; a sock takes up space. What is not yet publicly available is the data that turns those mechanisms into numbers: the actual internal clearance of a given cap at a given size, the true thickness of an issued work sock, the volume change over a specific shift pattern. Those values sit with manufacturers and with the wearers themselves. We are assembling them parameter by parameter, and until each one is measured we treat it as an assumption and label it as one.
The alternative — a fit recommendation that quietly reuses fashion-shoe tolerances for a shoe with a steel cap — would look precise and be wrong in the one zone where wrong hurts.
The fixable part
None of this requires new science. It requires the same thing that fit in any category requires: a measured foot on one side, a measured shoe interior on the other, and an honest account of what sits between them — the sock, the insole, the hours. For safety footwear the shoe side is, if anything, simpler than fashion. The lasts are few, the caps are standard components, and the manufacturers already hold the specifications. The data has never been assembled because no one has asked for it in that form.
Feet can now be measured with a phone. The remaining work is on the shoe side, and it is mostly a matter of asking.
The standard tells you the shoe will protect the foot. It does not tell you the shoe will fit it. Those are different questions, and the second one is answerable.
← All insights