The short answer
- Professional digitising runs $1–$4 per 1,000 stitches, most commonly $1–$3
- Whole designs typically cost $10–$85 depending on complexity
- A simple logo under 8,000 stitches is usually $10–$20
- Nearly everyone applies a minimum charge of $10–$20, which decides the price on small logos
- A badly digitised file costs more in thread breaks and rework than the saving
Digitising is the step between a customer's logo and something an embroidery machine can sew: stitch types, directions, densities, underlay and the order the machine works in. It is skilled work, it is priced in a way that confuses people, and it is where a lot of embroidery jobs quietly go wrong.
The 2026 market rate sits at $1 to $4 per thousand stitches, with most professional providers clustering at $1 to $3. Some advertise as low as $0.50 to $1.00, and effectively all of them apply a minimum charge. Whole designs commonly land between $10 and $85.
How do the published rates break down?
| Design | Stitch count | Typical 2026 price |
|---|---|---|
| Simple logo, left chest | under 8,000 | $10 – $20 |
| Medium complexity | 10,000 – 25,000 | $20 – $55 |
| Complex or large format | 25,000+ | up to $85 |
Notice that stitch count and price are correlated but not proportional. A 25,000-stitch design is not two and a half times the work of a 10,000-stitch one — much of the skill goes into decisions that do not scale with size, like how the underlay supports a fabric that will move, or how to sequence colour changes so the machine is not trimming constantly.
Why does the minimum charge matter more than the rate?
Because most of what a decorator digitises is small. A left-chest logo is frequently under 8,000 stitches, so at $1 per thousand the per-stitch maths gives $8 — below almost every provider's $10 to $20 minimum. The minimum, not the rate, is what you actually pay.
This has a practical consequence when comparing quotes. A provider advertising $0.60 per thousand against a $6 minimum can be cheaper on small logos than one advertising $1.00 against a $20 minimum, even though the headline rate looks worse in the other direction. Compare on your own typical design, not on the advertised rate.
It also tells you where to push back with customers. If a customer wants a logo digitised and then orders six shirts, the digitising is a real proportion of the job — and it is one-off. Say so, and price the reorder without it, because the file already exists.
You pay to digitise a logo once. You pay for a bad file on every garment that follows.
What does a cheap file cost later?
A poorly digitised design does not fail visibly at the quote stage. It fails on the machine, repeatedly, in ways that are easy to blame on the machine instead. Density set too high puckers the fabric and blunts needles. Underlay that does not suit the garment lets the design shift on knits. Poor sequencing means unnecessary trims and colour changes, which is machine time you are not billing for.
Signs the file, not the machine, is the problem
- Thread breaks clustering in the same area of the design every time
- Puckering on knits that does not appear on woven garments
- Small text filling in, or gaps opening between adjacent colour blocks
- Run time far longer than the stitch count suggests, from excess trims
- The design registering fine on one garment type and not on another
Any of those is worth a re-digitise rather than a fight with tension settings. The difference between a $12 file and a $30 file is one production run's worth of wasted time, and the good file keeps paying back on every reorder — which is why embroidery suits a recurring company store.
What should you send the digitiser?
Most of the delay and the rework in digitising comes from an incomplete brief, not from the artwork. A digitiser given a logo and nothing else has to guess at half a dozen decisions, and the guesses are where the file goes wrong.
Send all six, every time
- Vector artwork where it exists — the same rules as any other job
- Finished size in millimetres or inches, not "left chest"
- The garment — a piqué polo, a fleece and a cap all need different underlay
- Thread brand and colour numbers if the customer has a standard
- Placement, including how close to a seam or a placket it sits
- Machine format — DST, EXP, PES and the rest are not interchangeable
The garment line is the one most often omitted and the most consequential. A design digitised for a woven shirt will pucker on a fleece, because the underlay that stabilises one is insufficient for the other. Sending the same file to both is the usual reason a logo that has worked for a year suddenly looks wrong, and it is the kind of change a proof should record.
Size matters for the same reason. A logo digitised at 90mm cannot simply be scaled to 50mm — the stitch density that was correct at the larger size becomes too dense at the smaller one, filling in detail and stiffening the fabric. If a customer wants their logo on caps as well as shirts, that is two files, and usually two charges. Say so at the quote rather than after.
Should you digitise in-house?
It depends almost entirely on volume and on how much of your work is repeat. Software and training are a real investment, and digitising well takes months to learn rather than weeks. Against that, an in-house capability turns a two-day turnaround into an hour and removes a per-job cost that never goes away.
The rough test: if you are paying for digitising most weeks and your orders are mainly new logos rather than reorders, the cost is recurring enough to justify learning it. If your work is largely repeat business from a stable set of customers, the files already exist and outsourcing the occasional new one is cheaper than owning the skill.
Either way, treat digitising as billable rather than absorbed, and hold the file against the customer so the reorder starts from it — which is what our back-office management does with artwork. It is the same discipline as drawing a line on artwork time: work that takes a skilled person an hour should not disappear into the price of six shirts. The rest of the artwork archive assumes you charge for it.



