Nine Digitizing Mistakes That Ruin a Sew-Out
A stitch file can look perfect on screen and fail completely on fabric. These are the failures we see most, what causes them, and how to catch them before a production run.
1. Too much density
The most common problem by a distance. Packing in more stitches feels like it should mean better coverage; instead it perforates the fabric, creates a stiff board where the design sits, causes puckering, and breaks needles and thread.
Tell-tale: a very high stitch count for the size. A standard 3.5-inch left-chest logo usually lands between 5,000 and 12,000. If you're quoted 25,000 for a simple two-colour mark, something is wrong — see stitch count and pricing.
2. No underlay, or the wrong kind
Underlay is the foundation layer that stabilises fabric and gives top stitches something to sit on. Skip it and the design sinks into the material, edges wander, and coverage looks patchy.
Different areas need different underlay — edge-walk under satin borders, zigzag or tatami under fills, more on unstable fabrics. Applying one generic setting everywhere is a hallmark of automated output. Detail in underlay and pull compensation.
3. No pull compensation
Stitches pull fabric inward as they form, so shapes finish narrower than drawn. Without compensation, circles come out slightly oval and columns end up thinner than intended. On large designs the error accumulates visibly.
4. Bad sequencing
The order elements stitch in determines how much the fabric shifts and whether colours register against each other. Poor sequencing produces gaps where two colours should meet, excessive thread changes, and long jump stitches crossing open fabric.
Tell-tale: visible travel stitches across blank areas of the sew-out, or a colour change count far higher than the number of colours in the design.
5. Text kept too small
Lettering below roughly 5mm fills in and becomes unreadable. The mistake isn't usually the digitizer's — it's accepting a logo at a size where the tagline was never going to survive, and saying nothing. See small text in embroidery.
6. One file for every fabric
A file built for a stable woven shirt will sink into fleece and distort a performance knit. Caps need an entirely different build, stitched centre-out on a curved crown.
Tell-tale: nobody asked what garment it's going on. That question isn't small talk — it changes the file. See embroidery on different fabrics.
7. Auto-traced output passed off as digitizing
Automatic conversion inspects pixels and generates stitches. It has no idea what the design means, which element should sew first, or what fabric is involved. The results combine several faults above: excessive density, no meaningful underlay, illogical sequencing, no compensation.
Tell-tale: a price under about $5, delivered suspiciously fast, with no questions asked. More in auto vs manual digitizing.
8. Resizing a finished file
Density is fixed when a file is created. Scale it up and the same stitches spread over a larger area, leaving gaps; scale down and they compress into a stiff, puckered mass. Beyond roughly 10–20%, the design needs re-digitizing, not resizing — see opening and editing DST files.
9. No sew-out before production
Not strictly a digitizing error, but it's what turns a fixable problem into an expensive one. A stitch simulation on screen cannot show you puckering, registration or how the fabric behaves. Only thread on the actual garment does that.
Reading a proof properly
When you get a sew-out photo, check:
- Is the fabric flat, with no rippling around the design?
- Do colours meet cleanly, with no gaps or overlaps?
- Is small text readable, with open counters rather than blobs?
- Are satin edges crisp rather than wobbly?
- Any visible stitches travelling across empty fabric?
- How stiff is it? Flex the fabric — a board means too much density.
Any of these failing is a revision, and a decent provider fixes it free. If you're seeing several at once, the file wants rebuilding rather than patching.
Send us a file you're unsure about and we'll tell you honestly whether it's worth revising or starting again.
