Digitising is a craft, not a file conversion. Whether you buy it in or do it yourself, the quality of the brief decides whether you get one round or three, and how many rejects the job produces later.
The brief
Vector artwork. AI, EPS, PDF or SVG. If all you have is a raster image, say what it is and expect either a redraw charge or a compromise.
Final stitch size, in millimetres. Not "left chest", a number. A logo digitised for 90mm and run at 60mm loses its small detail and gains a density problem.
The exact garment and fabric. Brand, style, and fabric type. Piqué polo, jersey tee, 300gsm fleece, softshell, structured cap. Digitising decisions are fabric decisions.
Position. Left chest, centre back, sleeve, cap front, with placement measurements if you have a standard.
Thread brand and colour numbers. Madeira, Isacord or whatever you run, with actual numbers. "Dark blue" produces a colour conversation on the sew-out.
Stitch count ceiling, if you have one. If your price assumes 8,000 stitches, say so, and let the digitiser make the density decisions that keep it there.
Anything with small text. Flag it. Text under about 5mm cap height needs specific treatment or it fills in and becomes an unreadable smudge.
What good digitising gives you
| Property | Why you care |
|---|---|
| Correct pull compensation | Circles stay round, text does not distort |
| Appropriate density | No puckering, no thread breaks, no board-stiff patch |
| Sensible sew sequence | Fewer trims and colour changes, less machine time |
| Correct underlay | Stitches sit on the fabric rather than sinking in |
| Efficient stitch count | Lower cost per garment on every reorder, forever |
That last line is the one to internalise. Two files of the same logo can differ by 20 percent in stitch count with no visible difference. On a customer you sew for years, that difference compounds; see pricing embroidery by stitch count.
Always sew out before the order
A sew-out on the real garment, checked for registration, density, small text, pull distortion, and the appearance of the underside. Then approve the file. A file approved on screen is a file approved on hope.
Charge for it, and only once
Digitising is a one-off per design per garment type. Charge it as its own line, then never charge it again for that combination. Reorders that skip the charge are one of the easiest goodwill wins available, and they cost you nothing because the work genuinely is not there.
Keep the file with the customer, not in a folder
The reason digitising gets paid for twice is that nobody could find the file. Attached to the design and the customer record, a reorder in two years is instant.
The brief as a form, not an email
Everything in the list above works better as a structured field than a paragraph of prose, because prose lets people leave things out. The fields that eliminate a second round:
| Field | Why it decides the file |
|---|---|
| Final stitched width and height | Digitising is size-specific; scaling wrecks density |
| Exact garment and fabric | Piqué, fleece and performance knit each behave differently |
| Position | Left chest, cap front and sleeve are three different files |
| Thread brand and numbers | Prevents a substitute nobody approved |
| Stitch count ceiling | Gives the digitiser your cost constraint up front |
| Backing to be used | Changes the underlay decision |
| Small text present | The single most common cause of a redo |
If your quote already carries the garment, the position and the stitch count for pricing, the brief is mostly a print-out of the order rather than a fresh document. That is the point: the information exists, it just usually lives in three places.
Judge a sew-out against a checklist
A sew-out is only useful if someone inspects it properly. Six things, in this order:
- Registration. Do outlines sit on their fills, or is there a gap on one side?
- Small text. Legible at arm's length, or a fuzzy grey block?
- Puckering. Hold it flat and look across the surface at a low angle.
- Density. Is the fabric showing through, or is the fill so heavy it is board-stiff?
- Trims and jumps. Excessive jump stitches mean cleanup labour on every garment.
- Reverse side. Bobbin ratio and backing behaviour tell you how it will wash.
Approve the sew-out formally, attach it to the design, and note the run parameters that produced it. A reorder then reproduces a known result rather than repeating an experiment.
The economics of getting it right first time
A re-digitise costs the digitiser's fee twice, a second sew-out, and usually a delay that lands on a promised date. Worse, a barely acceptable file costs you on every single garment for years: thread breaks, trims, slower run speed, higher reject rate.
That compounding is easy to see once labour is measured. Clock on to embroidery runs, and the job costing report shows which designs consume hours out of proportion to their stitch count. A recurring order that runs 30 percent slower than it should is a re-digitising job that pays for itself in a month, and without the measurement it just looks like embroidery being embroidery.
Keep the file, the brief and the approval together
The file on its own is not enough. What you want on a reorder is the machine file, the native file, the approved sew-out, the thread numbers and the run notes, attached to the customer's design rather than scattered across a hard drive, an inbox and a machine's memory card. Printer's Friend keeps all of it on the design against the customer, so a two-year-old reorder quotes without the digitising charge and runs at the settings that worked. The pricing side explains why that reorder is the most profitable work in the shop.
Printer's Friend keeps the digitised file on the design so reorders reuse it, holds stitch count per position for pricing, and stores the sew-out approval with the order. Terms like pull compensation, underlay and backing are in the glossary.