It's true, isn't it? If all an organization had to do is bring one project in successfully, it would be a piece of cake. Most project environments, like R&D, IS/IT, product engineering, etc. are really multi-project environments.
The issue in this kind of environment is having clear priorities for the attention of resources so that the individual projects don't succumb to the lead-time-multiplying effect of multi-tasking. Project promises need to be protected from not only their own variation, but also from contention for resources. But before picking a tool to deal with the issue, one needs first to settle on an effective methodology.
The TOC/Critical Chain solution for multi-projects tackles the multi-tasking issue head-on. Appropriate synchronization of project launches helps to avoid having too many things on resources' plates without the futility of trying to level every resources' load. And Buffer Management helps to guide those resources in the fewer instances when more than one project is vying for their attention.
For more on the methodology, check out a paper I presented at PMI's Philadelphia Symposium a couple years ago. It can be found by clicking here.
Regarding tools, there are three out there that support this straightforward methodology, from ProChain, Speed to Market, and (my favorite and Gantthead sponsor) Scitor. The concepts are not rocket science, once understood, but they do involve a bit of paradigm shifting in the minds of many managers and executives, and therefore demand a bit of expert hand-holding for an effective implementation.
One of those paradigms . . . "The later you start most projects, the sooner they will finish."