Nobody reorders because the printing was wrong. They reorder because forty of the shirts do not fit, and that was decided weeks before printing.
Within an hour of the uniforms arriving, you know. Too many mediums, not enough XLs, and 11 people holding something they cannot wear. The printing was perfect. The order was wrong weeks earlier.
This is the most expensive avoidable mistake in corporate clothing, and it repeats because nobody writes down what went wrong last time.
Mistake one: asking people their size
People tell you the size they wore in 2019, the size they intend to be, or the size of the last shirt they liked. None of these is the size they need now, and none of them is deliberate dishonesty. It is simply how everyone answers that question.
Measure instead, or at minimum have a sample garment in each size that people physically try on before the count is finalised. A sizing session takes an afternoon. Reordering takes weeks and costs the whole order again.
Self-declared sizes: ordering to a list of them costs more than anything else on this page. A sample set in every size is a small, one-off cost that pays for itself on the first order it saves.
Mistake two: assuming sizes are standard
They are not, and they are not even consistent between garment types from the same supplier. An overall (KSh 1,500) is cut to go over other clothes, so its large is not a shirt's large. A reflector jacket (KSh 200, from 100) is cut generously for the same reason. A fitted corporate shirt runs smaller than a promotional tee in the same nominal size.
Ask for the measurements in centimetres (chest, length, sleeve), not the letters. Any supplier can provide them, and a size chart with numbers on it removes the entire argument.
Europe went as far as standardising it. EN 13402 labels clothes by body measurements in centimetres, chest or bust girth for most tops, with waist, height and other dimensions where they matter, because the old letter and number systems drifted and differed from country to country. Whatever system your supplier uses, asking for the centimetres puts everyone on the same footing.
Mistake three: ordering exactly the headcount
You have 60 staff, so you order 60. Then somebody joins in March, two garments are damaged, one is lost, and you are placing a second order for four items with all the setup cost of a first order.
Order a buffer, commonly 10 to 15%, weighted to the sizes that turn over most. It is cheaper as part of the original run than as a separate small order later, by a wide margin.
Mistake four: forgetting who wears it
Two practical things get missed. Women's cuts are genuinely different, and issuing a men's cut to everyone produces garments that technically fit and are never worn. And the person who works outside all day has different needs from the person at a desk: dust coats (KSh 1,000) for a lab and overalls for a workshop are not interchangeable, whatever the budget line says.
Mistake five: no plan for what happens when someone leaves
Branded uniform that walks out of the organisation is a small, permanent leak. Nobody wants to be the person demanding a dust coat back from someone who has just resigned, so it goes, and next year the count is short again for reasons nobody can explain.
The fix is a policy decided before the order, not an argument afterwards: either issue is logged and returned, or the garment is treated as consumable and the buffer reflects it. Both are workable. What does not work is having no position, ordering to headcount, and being puzzled every year.
The order that goes right
- Sample set first. One garment per size, tried on, counted properly.
- Record the sizes against names, in a file somebody else can find. Next year this list is the whole job.
- Buffer of 10–15%, weighted to the middle sizes and the extremes you were short of last time.
- Approve one finished sample before the run. Not a mock-up on screen: a real garment with the real branding on it.
- Check the fabric against the work. Heavy cotton is miserable in a hot warehouse; light polyester is useless outdoors in cold.
- Agree the reorder route now, so March's four garments are a phone call, not a project.
What to spend the difference on
Where a budget is tight, spend it on fit and fabric before the branding method. A well-fitting plain uniform with a modest embroidered logo is worn every day. A poorly-fitting one with an elaborate print sits in a locker, and a uniform nobody wears has a return of exactly nothing.
On which branding method to choose once the sizes are settled, the comparison of the four covers where each one stops making sense. For uniforms specifically, embroidery is usually the right answer for the reasons in why prints crack.
Sources: EN 13402 size designation of clothes (Wikipedia) (read 14 Oct 2026).
Common questions
Can I order a sample before committing?
Yes, and you should. A sample garment with your actual branding on it, in your actual fabric, is the only reliable way to approve a run; a screen mock-up cannot show you how the logo sits on a curved chest or how the fabric behaves. Where sizing is uncertain, ask for a size set instead of a single sample.
What is a realistic lead time for 60 uniforms?
Plan on a couple of weeks from approved artwork and a confirmed size count, and understand that the count, not production, is what usually delays it. Orders held up at our end are rare; orders held up waiting for a department to return its sizing list are the norm. Start the sizing before you start the artwork.
Can we add new staff to an existing design later?
Yes. Artwork and, for embroidery, the digitised stitch file are kept, so a later order does not repeat the setup. What changes is the per-item cost, because four garments carry a different economy from sixty. This is precisely why the buffer in the original order is worth having.
Do you supply the garments or do we?
Either. Supplied garments come as one price with the branding included, which is simpler and usually cheaper at volume. Bringing your own works too, and is common where an organisation has a specified fabric or supplier; bring one item first so it can be tested, because not every fabric takes every branding method.
