This project reminds me of Jared Spools $300 million dollar button example. Our products initial research participants were single user project managers. As soon as we started to expand to larger organizations, our model fell apart. We had created a process that made perfect sense from an engineering perspective, but lacked insight and affordance to the needs of our users in the context of their day to day jobs.
The fix was simple...once we understood what the problems were. This process reinforced the need to visit with our users to determine not just what jobs need to be done, but who will be doing them and the constraints they are working with. There is no substitute for observing the user in their environment or better yet, using the software under similar conditions to do their job.