First, I was going to build my own - I used that same greymatter blogging engine that powered Portrait Adoption and EMG for so long, and laboriously built a few beta-test sites. They... were extensive to build and maintain - far too involved to do for free, and it hadn't occurred to me then to charge for them. When Portrait Adoption and the giftshop were moved to php, I hired a programmer to build me a site that would let an artist upload their artwork, have me approve print files, and offer them little custom-made shopping pages that would order from me and let me pay the artists anything above the base pricing. The programmer worked fast and did a fine job (I hired a number of programmers about that time...), but... the flow of the site was clunky. I still had to do a lot, there were some bugs (normal in beta testing), and my profit margin was minuscule. I let it run for a while, with a handful of beta testers, and sold one print. "This," I said to myself, "is not a good plan."
I let those beta testers keep their sites until the domain expired, some three years later, but never took the project further, and of course, just last year, I pulled the plug on every product but prints.
I do still provide prints for a select dozen or so people. My prints are - and I'm not exaggerating - without compare. They are high quality, with a fast turnaround, and exceptionally carefully done. They are not nearly as cheap as other printing services, because I use only the highest quality papers and inks, and I've learned many (expensive) lessons about how to price so I don't screw myself. I do prints for a few locals, a few not-so locals, and a few big names - people who already know how to provide good quality print files, are good to work with and have, over the years, done good consistent business with me. Most of them don't order huge quantities at once, but they are, generally, steady.
I don't make a ton with these services, but they do keep my printers in action (which is important - you don't want printers of this type sitting around getting clogs!), and they pay for themselves.
Ironically, I've completely automated this service through the lilypad (the EMG backend) in terms of uploading, managing and ordering prints, but none of my regulars use it... preferring to keep using ftp and email. Go figure! At least they know what they're doing and asking for, so it doesn't waste too much of my time.
I see no real reason to stop providing existing clients my print services, but I'm not terribly interested in offering them generally. Possibly in the future, this service (and maybe a related Limited Edition service that uses certificates like PA does) will become publicly available. But for now? Nah. Not that I hate doing prints, but there isn't enough reward to go through the learning curve that comes with new clients. This is off my radar entirely right now.