At a large HMO I used to work for, they used to have a tough re-image policy.  I believe that if a problem took more than two hours to solve, they would just have the desktop support person re-image the machine.  This approach did lead to plenty of problems, lost user data, etc.  But, if managed properly it probably could have cut down on user downtime and support costs.

  1. User data – My current approach is to backup all user data to the server anyway.  Sometime users will save files outside the standard folders (i.e. My Documents).  This HMO also had a clever script that would locate all supported data files (i.e. – .doc, .pdf, .pst, etc.) and back them up prior to the image drop.
  2. Non-standard software – Once drawback to this method is that IT didn’t necessary have images that included ALL of the software required by all the departments.  One group required SAS tools that weren’t in the base image.
  3. Software updates – This approach would work best if the images were updated with the latest Windows updates, etc.   Otherwise you need to re-run all the patches each time you drop an image.
  4. User profiles – On the one hand you could keep the user profile on the network, but on the other hand sometimes a corrupt user profile is the cause of the problem.

I haven’t tried this method at the small business level, but I think that it would be very difficult to reduce support costs by taking this approach..