The short answer
- Weight and fibre content say almost nothing about how a garment behaves
- Test printing, washing and fit before putting a style on your price list
- Consistency between deliveries matters more than any single sample
- A cheap blank that prints badly is not cheap
- Test with your own process, not the supplier's photographs
Choosing a blank from a catalogue is choosing on three numbers: weight, fibre content and price. None of them predict whether the garment takes ink cleanly, holds its shape after five washes, or fits the way the customer expects.
Testing takes an afternoon and a handful of garments, and it prevents the specific failure that damages a shop most — recommending a blank that disappoints on a large order.
What the spec hides
| The spec says | It does not say |
|---|---|
| Weight in gsm | Whether the knit is tight or open |
| 100% cotton | Ring-spun or open-end; combed or not |
| Colour name | How much it varies between lots |
| Size chart | Whether the garment matches it |
| Preshrunk | How much it still shrinks |
Yarn quality is the largest omission. Two 100% cotton tees at the same weight can behave completely differently depending on how the cotton was spun, and the smoother of the two will print noticeably better.
Weight tells you how heavy it is. Nothing on the spec sheet tells you how it prints.
Fabric
Handle it and look closely
- Surface smoothness — a rough knit breaks up fine detail
- Knit tightness held up to the light
- Whether it feels soft or stiff, and how it drapes
- Slubs, thick spots and visible irregularity
- Stretch, and whether it recovers
- Colour depth and evenness across the panel
Surface smoothness is the property with the most effect on decoration. A rough or open knit lets ink sink into the texture, so fine lines and small type lose definition — the same constraint discussed in typography on garments.
Check colour evenness across a single garment as well as between garments. Panels dyed unevenly show most under a light print, and it will be blamed on your printing rather than on the blank.
Construction
Look at the seams
- Stitch density and whether any threads are already loose
- Side seams versus tubular — tubular twists after washing
- Shoulder taping, which holds the shape
- Collar ribbing — does it recover when stretched
- Hem finish and evenness
- Label type: sewn, tearaway or printed
The collar is the fastest indicator of a garment's life. Stretch it firmly and let go — a collar that does not return will look worn after a few wearings, and the customer will remember which shop supplied it.
Tearaway labels matter to anyone doing branded work. A blank that can be relabelled opens up private-label orders, and that is a commercial capability rather than a detail.
How it decorates
This is the test that matters, and it must be done on your own equipment with your own settings.
Print a real test
- Fine detail and small type at your usual minimum
- A solid block, to see coverage and opacity
- A white on dark print, the hardest case
- Halftones if you run them
- A cure check at your standard dryer setting
- An embroidered sample if you offer it — how does it handle a hoop
On polyester and blends, test for dye migration specifically, and leave the sample for twenty-four hours before judging. A blank that migrates under your normal cure settings is a blank you cannot sell to teams in bright colours.
Note anything that required a process change. A garment needing a lower cure, a different platen or an extra flash costs you time on every job, and that cost belongs in the comparison rather than in the price.
Wash performance
Wash the decorated samples the way a customer will, not the way a laboratory would.
After ten cycles, check
- Shrinkage measured, not estimated — before and after
- Colour fade in the fabric itself
- Print condition: cracking, lifting, fading
- Shape retention through the body and collar
- Pilling on the surface
- Seam twist, particularly on tubular bodies
Shrinkage is the one customers notice most and complain about loudest. A garment sold as preshrunk that loses a size in the first wash produces a complaint you cannot answer, and measuring it yourself is the only way to know — the method is the same as ordinary wash testing.
Consistency
One good sample proves very little. What matters commercially is whether the tenth delivery matches the first.
Watch across deliveries
- Colour between dye lots, compared side by side
- Sizing between production runs
- Fabric hand — does it feel the same as it did
- Construction changes made without notice
- Stock availability across sizes and colours
Silent specification changes are a real risk on long-running programmes. A garment reformulated between seasons still carries the same style number, and the first you know is a customer comparing this year's shirts with last year's — which is why samples from each order are worth keeping alongside the receiving records.
Running a comparison
Test properly
- Two or three candidates at once, same day, same settings
- The same design on all of them
- The same wash cycles
- Score against a written list rather than an impression
- Include cost per garment delivered, freight included
- Have someone else look without knowing which is which
Test the sizing against the published chart as well. A garment that runs a size small is not a fault, but it is something you must know before a hundred people order from a size guide, and the discovery is far cheaper on a sample than on a closed store.
Blind comparison removes the pull of the cheaper option. Shops routinely find the mid-priced blank wins on every measure that reaches the customer, and the difference in landed cost is smaller than the price list suggested once freight and minimums are counted.
Write the result down and keep it with your product data. A tested, documented range is what lets anyone in the shop recommend a garment confidently, and where that range lives in the catalogue your quoting and stores draw from, the tested choice is the one that gets sold rather than whichever style somebody remembered. The rest of the suppliers archive assumes blanks were tested before they reached a customer.


