

The one part of digitizing you control is what you send. Here's how to prepare artwork properly, where to find the good version, and what to do when it doesn't exist.

Most of digitizing happens after you've handed the job over. The one part you control completely is what you send, and it decides more about the result than people realize.
Send clean artwork with a proper brief and the file comes back faster and better. Send a photograph of a business card and you've added a rebuild to the front of the job, before anyone has thought about a single stitch.
Here's how to prepare artwork properly, where the good version is usually hiding, and what to do when it genuinely doesn't exist.
There are two kinds of artwork, and the difference is the whole subject.
Vector files describe shapes mathematically. AI, EPS, SVG and most well-made PDFs. You can enlarge them endlessly and the edges stay perfectly crisp. A digitizer can work straight from them.
Raster files are grids of pixels. JPG, PNG, GIF, most things you download. Usable if they're big and sharp, useless if they're small or blurry.
The quickest test costs nothing: open the file and zoom in hard. If the edges stay clean at 800 percent, it's vector. If it turns into blocks, it's a picture, and the size it is now is as good as it will ever get.
People send what's on their phone. The better file usually exists somewhere else.
What isn't a source: a website favicon, a screenshot of a web page, a photo of a signboard, or a logo lifted from someone's social media profile picture. Those are all small pictures of the logo, not the logo.
Prepare artwork this way and a digitizer can start the moment it lands. A few minutes of tidying saves a day of back and forth.
Common, and not a disaster. Say so up front rather than hoping nobody notices.
What happens is that the artwork gets redrawn before digitizing starts, which is a real piece of work and a real cost. Some digitizers include light redrawing, some charge separately, and almost all of them will do it faster and cheaper if you tell them in advance instead of sending a blurry file and waiting for the question.
Send the biggest and sharpest version you have, even if it's imperfect. A 2000 pixel wide JPG is far more use than a crisp 200 pixel one. And if all you have is a physical item, photograph it flat, straight on, in daylight, filling the frame. That's genuinely usable. A phone photo taken at an angle in a dark room is not.
This rebuilding time is one of the things people are surprised to be paying for, and it's covered in why embroidery digitizing is expensive.
Artwork alone isn't a job. Send these at the same time and you remove almost every reason for someone to come back to you:
Every one of those that's missing is a message and a wait. Our post on how long embroidery digitizing takes makes the point that queue and clarification, not the digitizing itself, are what stretch a job to four days.
Six ways people prepare artwork badly, all of them well meant:
Worth knowing before you go hunting for a better file, because sometimes the file isn't the problem.
A photograph can't be embroidered as a photograph. It has to be redrawn as flat shapes, and what comes back is an interpretation rather than the image. Very fine detail has the same issue: a crisp vector hairline is still a hairline, and thread has a thickness whatever the source file says. Gradients, glows and soft shadows don't exist in thread either.
So a perfect vector doesn't guarantee a perfect stitch-out. It guarantees the digitizer is deciding rather than guessing, which is a different thing. What survives the crossing and what doesn't is covered in digitizing a logo for embroidery.
Being sent a logo isn't the same as being allowed to reproduce it. For normal corporate work the client owns their mark and there's nothing to think about. For anything else, a short message confirming they have the right to use it costs you nothing and matters if it ever comes up, because you're the one who stitched the garments.
Once you've chased down clean artwork for a client, that file is worth money. So is the digitized version, the thread numbers and the sizes you used.
A year later the same client wants caps, and the difference between quoting confidently and starting from scratch is whether any of it can still be found. Keeping it with the client record rather than in a folder somewhere is exactly what EMBBA is for.
All of this applies to you twice over, because you're both ends of the conversation. The habit worth building is refusing to start until the artwork and the brief are right, even when the client is in a hurry, since every minute saved at the front reappears as two at the back.
Our Digitizing Mastery course covers artwork preparation as part of the workflow rather than as an afterthought, along with the machine-side judgment that decides what the artwork can become.
Vector, so AI, EPS, SVG or a properly made PDF. Failing that, the largest and sharpest PNG or JPG you have. There's more on formats in our guide to embroidery file formats.
Yes, usually, by redrawing it first. That's extra work and it's slower, so expect it to affect the price and the turnaround.
Not directly, but poor artwork leads to guesswork about edges and detail, and guesswork tends to produce heavier files than necessary.
Bigger is better and there's no real upper limit. As a rough floor, aim for something at least a thousand pixels across the logo itself, not the whole page it sits on. Below that a redraw is almost certain.
No, send everything and let the digitizer tell you what won't hold. They'll suggest the alternative, and you keep the decision.
Prepare artwork before you send it and the whole job goes better. Find the vector if it exists, tidy it, name the colors and fonts, and send a complete brief with it. When there's no clean version, say so at the start. It's the cheapest thing you can do to improve a result you otherwise have no control over.
The Academy turns articles like this into proper courses, with video lessons, assessments, and people to learn alongside.