The challenge of user file proliferation

Where are all the files that your users create ?

On their hard disk ? On some storage somewhere ? In the cloud ? the answer, is more than likely, all of the above – what is this about and can an enterprise hope to manage the great morass of files that user create.

Lets define the problem

It can be seen from many perspectives, however the way I look at is that user files proliferation is like entropy, unless you expend effort to keep them organized, they tend towards disorganization and randomness.

Why this happens is that we all get busy and making sure that every files goes where it should requires a level of obsessiveness that is bordering on compulsive – and none of us want to consider that now do we ?

Modern operating systems with their file indexing and search can make it better (or worse ?) by helping the use find the files, but this doesn’t actually address the fundamental point that on the users on machine is not where these should be stored if they have business value to the enterprise.

The principles of storing user files

The basic principles that should be applied are;-

  • The user should require minimal (or no) intervention to ensure that the file is stored where it should be stored
  • The retrieval of the file by the user should, similarly, be simple and straight forward;
  • There will be many places where the file may reside based on the importance or other attributes e.g. security classification, of the file to the enterprise.

This last principle needs a bit of explanation – if you have created a letter of offer to a prospective employee, the enterprise policies may require that that be kept for 7 years and be stored on long term storage inside the enterprise. A user is not going to be able to parse all the rules, and ensure compliance on their own – so what I am suggesting is that the infrastructure needs to be assisting each and every user in completing the task to the satisfaction of the enterprise.

Ideal vs Real world situation

In an ideal world, when the user presses “save” and the file gets written to disk, an agent takes up the task of analyzing and categorizing the file to figure out the storage category appropriate for the file then puts it there.

Sounds complex and it is.

Real people in the real world just want to complete the task and get onto the next one, so what happens in most enterprises is there is a simple, and somewhat binary approach. Any storage provided by the enterprise, meets all but the most stringent needs, anything else is “not permitted”. This includes the users own device – with whole of device encryption & key escrow being the common manner in which this is dealt with.

How do you solve this ?

Maybe there is no right solution here – every user is going to be a bit different, however I can see 3 key strategies to ensure some level of compliance;-

  • Education – like we do with Cyber security – SPAM, phishing etc. we could add this to the list of things that we ask folks to be reminded about every year or so.
  • Simplification – try not to get too complex in the rules that are applied – in my experience, every line in the set of rules reduces somewhat the chance of the user getting it right
  • Infrastructure based – put services into the back end to analyze and classify what the user puts up – and advise the user of changes needed, or better yet do the changes for them.


In conclusion, every enterprise will have a sea of user files with many of these probably not in the most appropriate type of storage based on the importance to that enterprise. Setting up a service as an assistant to users in curation of these files is a key technique in getting the entropy laden user file storage under some sort of control, and you never know, you may reduce or delay the next big storage upgrade.

Thanks for reading.