Inside each element folder, you will typically see only two things–the latest version of a file (or if it’s a multi-document item, a LATEST folder), and a folder called “OLD/OTHER/BACKUP.” Occasionally, there will be a NOTES or WORK folder as well. If a document is moved to a new version, the other version is moved to the OLD folder and the new version resides as the latest.

I frequently work from home, in which case I change the name of the LATEST folder to LATEST HOME WITH KATIE, so people know not to make changes in those items. When I work from home, I keep my files in my Dropbox, so that if they desperately need to access them from the office, they can.

When a year is complete, the entire year folder is dragged to the “PAST YEARS” folder, and the substructure exists as it always has.

“By name” folders are kept under “STAFF FILES” on the server and are typically only used for items that don’t fit anywhere else and wouldn’t mean anything to other employees.

Documents that go from project to project are kept in master folders outside of the yearly folder system. For example, I keep data and merge shells, as well as document templates and automator workflows, in a folder called MASTER SCRIPT ELEMENTS.

I suppose a lot depends on what type of work you do. Our projects are very modular, so it’s easy to keep them organized, and we impress on everyone who comes into contact with our server that we expect the structure to be maintained.

But I do believe that almost anything can be organized well if you take it from the perspective that a stranger should be able to log in and find a specific item by making choices from among the available folders at each level.

Good luck!

But more to the point, the date in the file name isn’t necessarily a create date. It might a publication/issue date, a revision date, or some other key date that isn’t reflected in the file’s properties.

Sometimes, too, it’s the only thing that’s distinctive about the file name, e.g. Newsletter_090105; Newsletter_090204; Newsletter_090304; etc. and so the date is necessary, given that identical file names within a folder aren’t permitted (or useful).

Re naming folders, create a structure then mentally try to file everything in one of the folders. If there is any confusion, try another structure; repeat until it works. Fewer folders the better. And don’t use the organisation’s business structure as a filing structure, it rarely works.

The other thing I do is keep a folder labeled zArchive (to sort it to the bottom of the list). Every ‘old’ version gets thrown in here, so folders only contain the latest versions, but the old ones are available if I really need them. Never, ever do the reverse of this i.e. create a “Latest version” folder – it will end up like the moving box a friend found labelled “urgent – do immediately – April 2003″!

My final suggestion would be to browse for blogs or websites on records management and look for help there. They are the professionals.

While there are obviously differences in system between the areas, this practice does work better for us than some arbitrary system imposed from on high. That’s because each area can adopt a system that works best for the nature of its work and workers (and their levels of fanaticism) and because there is ownership of the system as well as personal motivation to maintain it.

If I have a gripe re my workplace’s filing practices it would be that not everyone understands/appreciates the importance of using a reverse date system (y/m/d) in file names to ensure logical sorting. So we do get a bit of d/m/y in some parts of the system (this is in Australia where d/m/y is the standard sequence for dates).

Julia’s question is a really important one. In my own work area, for example, I *frequently* need refer to or reuse work and files from previous years, and as I work there longer so will the number of years I’m referring to increase. To a lesser extent, colleagues would refer to my older files too. So for me the “archives” are actually working reference files and need to stay available. But there would be other work files that can quite easily be archived to DVD or some other medium.

Perhaps one solution for the intrepid deleter of old files would be to archive everything to another medium, but to leave in the original location a spreadsheet indexing what was archived. Not dissimilar to the spreadsheet that you’d create to itemise what was in the boxes of physical files that have been sent to a storage unit or paper archive.

Any suggestions?

Also with the folders for different groups and permissions set; mapping the drive to the folder they have access to reduces clutter. Also when people leave a job IT should clean their network space not just their computer.

Having everything indexed in a spreadsheet is nice, but someone has to be fanatical about updating the spreadsheet, too. It’s extra work, and if it doesn’t get done, the spreadsheet is worthless. Better to have a system that everyone understands and follows.

********850 to find all your 850 files. You can find them EVEN IF YOU MISFILE.

The * break down this way:

****/**/**
year/mo/day

Here’s how it works.

The main folders are 100, 200, 300, 400, 500, 600, 700, 800, 900. You could give them names 100 = Computer, 200 = Finance, 300 = Research, etc. Whatever. Whatever your needs are.

Within each folder, there are other folders. I will use folder 800 as an example:

810, 820, 830, 840, 850, 860, 870, 880, 890.

So let’s say that Folder 800 = Publishing. Folder 810 could be Vendors (I’m making this up on the fly, so lousy examples, forgive me)

810 Vendors
820 Suppliers
830 Clients
840 Authors
850 Conferences
860 Digital
870 Software
880 Paper Manufacturers
890 Press

Within each folder, you can break it down some more. Say for Folder 810, you could have 811, 812, 813, 814, 815, 816, 817, 818, 819.

You get the picture. This system expands to meet the needs of the largest company in the country, yet can be used by a one-person shop depending on how deep you want to go.

Now….HOW TO USE IT.

Using Folder 850 as an example. You have a file that contains your registration for an Adobe conference in September 2009, and you filled it out today, May 8, 2009. Here is how you would name your file.

20090508850 Adobe Whatsis Conference 9_2009

or

20090508850 Adobe Conference Change 9_2009 ver 1

You have the date properly formatted up front for easy indexing, and the last three digits of the date — here it’s 850 — are the folder number, according to YOUR system. You can use this system digitally as well as on paper folders in your drawer without any duplication of effort and you maintain business privacy.

If you are a small two-five person shop, your folder numbers would be well known to your team: 850, 110, 446, whatever. You can name your files this way and no one would be able to tell the contents from seeing it on a desk except those who have a need to know.

In the case of the medical insurance company mentioned above, they used it this way in an end-tab system:

20071230R323

2007 was the date the patient first saw the doctor
12 was the month
30 was the day
R was the first initial in the patient’s last name
323 were the last three digits of the patient’s social security number.

The only important part of this file name is R323. Armed with this they were able to locate the Rs in 50,000,000 files, and then locate the patient by social security number. The likelihood of a patient whose last name begins with R and whose last three digits in his SSN are 323 are a billion to one.

They were able to locate files with this system nationwide within two minutes, and maintain privacy.

Folders named with job numbers (which can be customized however you want – client name, start date, etc.) and a shared Google spreadsheet would solve 95% of everyone’s problems.

I freelance and work in all kinds of environments. There are tons of fancy solutions out there, but simpler is always better.

I got lots of good information for document management at http://tinyurl.com/iso15489

Luckily every modern file system (NTFS, Mac Journaled, etc) has permissions feature and they should be used judiciously. users in the “marketing group” shouldn’t have access to the “engineering group” or “financial group” and vice versa, people from different groups should communicate with users in those groups for the information they need, this way everyone knows what information everyone is using.

better still would be to implement some kind of version control system, even though version control systems were originally designed to organize software development. A centralized version control system like Subversion can be used to keep track of all the files on the drive, who accessed it, and each file’s complete history. it will work with any type of file, and has the added benefit of being entirely OS independent, as it operates on the TCP/IP protocol. using Subversion you can have a shared drive which any OS can access. the workflow is a little different than simple drag and drop, but easy-to-use tools have been developed to make the process just as easy. http://subversion.tigris.org