I think I'll explain our system a little more.

On our job server, each project has a folder with a job number, then a brief description. Often the client name is referenced in the folder name:

4287 - JonesCorp Brochure

Then in that folder we have 4 (or so) subfolders to keep things organized:
- copy
- images
- proofs
- working files

"Copy" contains all the text, "Images" contains all the image support for the project, "Proofs" contains versioned PDFs labeled r# to indicate how far along the revisions have come (in an effort to flag clients looking at r8 ? Revision #8), and "Working" contains all the layered and in-progress files, including alternate designs for a project, which is different from revisions, which are tracked in the Proofs folder.

When it's done, a folder named "Final" will be added, containing all the final, collected, press-ready files for the project.

This way, when your co-worker is out sick and you have to take over his project, you aren't bombarded with a bunch of files that you have no idea what to do with. But the key here is to have ONE system that everyone in the office agrees on and abides by. We didn't start naming all the proof PDFs with the "jobnumber-jobdescription-revisionnumber" format until a month or two ago. (ie: "4287_jonesbroch_r3.pdf") However, we have been on the project folder structure mentioned above for about a year. Consistency is the key, though, and what works for krakeman's business may not work for mine, and vice-versa.

Finally, our job server has a folder named ARCHIVE, which is where we put finished jobs, and eventually archive everything onto disc like I mentioned a few posts ago.