So Fiona, I take it that your organization has never been sued, or faced an ediscovery situation yet? did you know that in some states like California, judges instruct juries to ASSUME that any information that cannot be produced was deleted on purpose? The liability alone should make your business case.

To respond to your questions specifically.

1. It will work, but first you have to stop the bleeding. Luckily it is only 50 or so users, so you can train them easily on the new processes and new systems. My advice would be to lock that volume as read only once you get a new system.
2. I think you can expect significantly improved findability of information, greatly shortening the time it takes to find the right file or document. Likewise if you audit those records, you can most likely reduce the amount, cleaning up the signal to noise ration on that content. An analogy I give is that if done right, it should be as easy to find things on an internal system, as it is to search for facts on Wikipedia.
3. This is variable, but what you can do is take a look at 5-10% of the system, and estimate how long that will take (so 25 - 50k records) and extrapolate from that sample size. Still, there are some hidden variables such as change management and process control you'll need to factor in. my quick advice is however long you THINK it will take- double it. It will set expectations, and give you room to breathe.
4. What you are doing is incredibly important, and yes it won't be appreciated until that organization is sued, or audited- and they have to produce records from your new system. In fact, you have two methods of making the business case internally- carrot and stick. The carrot is all the nice things, speed, accessibility, better client service, and faster collaboration. The stick of course is what happens if you don't change things, and something occurs.

Last but not least, make sure that your organization is fully committed to process improvement, and that the people are ready for it as well. With a good communication and training plan, and clear business objectives, you give yourself the best probability for success.

If you have not yet chosen a vendor, AIIM has a wonderful guide of companies you can speak with. Since you are in the UK, if you would like a reference please let me know. There is a company I have worked with that has tackled similar problems and I think would be able to help you.

Fiona,

Some insight on several of your questions. I been consulting for 20 years and I think many companies do not see the value ($$) in organizing information like fileshares. While Daniel makes a good point about litigation, chances are this will not impress them or sway them (even companies that are in litigation don't always change their habits.)

One thing to suggest is to do a scan of the fileshare (there are even free tools for this) and identify duplicate files and older files past an agreed upon date (there may even be file types no longer used which you can idenitify by their extension). Once identified, try to get them to delete these files. My experience is that around 50% of the files could go away. This will reduce the overall load, but you still need to come up with a new organizing principle, like Daniel suggested.

As to oganization, look at the possibility of SharePoint and creating libraries. The company may own a copy of SharePoint if they have Microsoft servers - so this would be free. SharePoint libraries could be created to organize the files and maintain them. When successful, the basic SharePoint can be upgraded to a more advanced version that includes records management. (Which means that you have to develop a records retention plan...).

Bud

Fiona, I've been in your shoes before and sort of live in them at my present assignment now. Perhaps an analogy might work for this organization: if you knew something were wrong with your health, would you go straight to the surgeon and have him or her start cutting before time (and money, even if in the UK you aren't paying it unless in private care--it still costs something to do the diagnosis) were taken to analyze the problems, form a diagnosis, mitigate things as best a doctor can BEFORE you determine what type of surgery you should have?

I often had potential customers say, "can't you just tell me what software to buy" before I even had an idea of what was the exact problem to fix, the size of the problem, and what will be the needs going forward.

I see there's been a lot of time since you first posted so I'm hoping this has sorted itself out since, or you are still in negotiation with them. They may just be determined to take advantage of someone who is a new graduate. You are following your instincts correctly and you will have to make a choice as to whether the job warrants pursuit.

I found from 30 years in professional information specialization (libraries, archives, electronic database publishing, and other consulting on database work) that you need to distinguish between giving the customer what he/she asked for vs. giving them what they really want. It's very often not the same thing.

Good luck.