As we near the product release date for Exchange 2000 Server, many of you are leading or involved in projects to pilot the rollout of this promising new technology. At a recent demonstration of Microsoft Project Central for sharing team project information, the presenter opened a sample template file in Project 2000 that outlined an Active Directory (AD) deployment. Thanks to such open information sharing within our industry, we all stand a good chance of a successful Exchange 2000 rollout. And your chances of successful deployment are much greater if you're doing pilot testing or conducting a thorough network architecture analysis as part of some type of project management methodology.

The importance of successful project management was re-enforced recently when I read news reports of a $38 million financial computer system project at my local county government that is being considered a failure. Apparently, three things caused the project's failure: The people involved didn't fully consider computer and business procedures; they brought part of the system online before it was ready; and they spent the rest of the budget fixing the problems that followed. Now the group is spending an extra $3.7 million to shut down the project. Of course, some project team members are defending the project, claiming success on the limited system that they did bring online. These people state that the problem was simply the original budget estimations.

Will your Exchange 2000 Server project be considered a success if you merely roll out messaging for a limited set of users, fail to deliver training, deploy a limited feature set—and then claim that you didn't get enough funding? Most definitely not. Even a well-designed, well-built system is of little value without the accompanying training and support for administrators and end users.

The picture that accompanied the newspaper article showed rows of expensive midrange computers sitting idle. Failures of this magnitude are less common in PC server-based deployments, thanks in part to standardized solutions that draw on a wider knowledge base and skilled labor force, but good project management skills are crucial, nevertheless. For example, a good project manager follows a methodology or solutions framework. The framework establishes what needs to be done (e.g., analysis, planning, design, test build, deployment, and maintenance) and in what order. Much of this approach is common sense, such as the fact that the developers can't come up with solutions before a thorough analysis, and developers must thoroughly test and pilot these solutions before production deployment.

The demands of tight deadlines and limited funding test the project manager's ability to focus on key success priorities. One reason the application service provider (ASP) market is so hot and the service provider business model makes sense is the area of project management. In the IT business, we're in a constant state of upgrade and change. If we don't manage these changes well, we put the core business at risk. The service provider must successfully deploy new technology or risk losing the customers they serve or host. Companies that have chosen to outsource their messaging system are in a position to watch the service providers to see who will be the first to successfully manage projects and deploy fully functional systems. And all of us will benefit if these companies publish the information and best practices from these successful deployments.