Life as a scientist, especially for postdocs, revolves around a project, or more likely, several projects. Graduate students receive help managing projects from an advisor, mentor, a committee, and sometimes even a department of concerned faculty. In contrast, we postdocs are expected to manage projects on our own. If we're lucky, a skilled mentor will offer guidance. But we can't count on that. It is therefore essential that postdocs learn to be self-reliant project managers.

This year the University of California, San Francisco (UCSF), Postdoctoral Scholar's Association chose to highlight project management in our first Practice of Science seminar. A trio of career project managers who double as consultants and educators were invited to lead the seminar. Ray Sanchez-Pescador, a certified project manager with Polaris Management Consultants, co-teaches a course at UC Berkeley with Genia Guilbert, another of the panelists. A third panelist, George Ksander, senior project manager at Genentech, introduced concepts and benefits of effective project management from the front lines of a biotechnology company.

Perhaps it is obvious, but a critical first step is defining the project. Ksander admitted his admiration for the aerospace engineers who build military technology: "Designing and building a B2 bomber, now that's a project!"

A project must have a tangible outcome and a definite timeline. The Project Management Body of Knowledge (2000, Project Management Institute) defines a project as a "temporary endeavor undertaken to create a unique product or service." For postdocs, the unique product may be the publications or patent that help them land a full-time position. Sanchez-Pescador emphasized that it is important to distinguish between project, product, and process. Developing an assay is a process that might be a useful resource for a future project.

Most postdocs are project managers, but few admit to being natural managers. Born planners are easily recognizable: They are constantly on top of progress, what gets done when, and fiercely goal-oriented. But others are more likely to grab a pipettor with one hand and start sketching on a piece of paper towel with the other--just to get started--even if they're not exactly sure of their plan.

Many scientists think that the planning phase of a project begins and ends before the project starts. Once the project plan is written, all future effort is dedicated to doing experiments. We often feel guilty if we're not executing some sort of plan. However, Ksander argued that planning should be a continuous process that lasts throughout the project. If done well, planning should take 20%-30% of the total project time!

Sanchez-Pescador cited a 1998 survey of information technology projects ( Chaos: A Recipe for Success) that found nearly 30% of projects were cancelled midstream. The study found that most successful projects were often twofold over budget and took twice as long as planned. The numbers may be even bleaker in academia. Of course, not all projects are going to be winners, but failure rates can be minimized with careful risk-assessment and contingency planning.

Successful projects depend upon thorough and continuous planning and goal reassessment. The ability to envision many pathways to a stated goal is a crucial project management skill. Pathways need not be linear. In fact, they are often composed of multiple, parallel, and interconnected steps that can be assigned relative probabilities of success. The rate-limiting step is a focal point for risk management. Failure here will often cripple a project, unless contingency plans have been worked into the project.

Professional project managers see projects as a formula from which the desired endpoint in time and cost can be calculated. In addition, a schedule of task execution can be represented graphically in what is called a Gantt chart. You may have seen one of these charts, as contractors planning lab renovations often use them. However, few postdocs have been taught to create such a chart for experimental lab projects. For many postdocs, a project timeline is often little more than a sense that we should be publishing n papers per year.

But actually, a project timeline section is part of most federal science grants. In basic research labs, rarely are these timelines followed, or are the goals explicitly expressed to lab members. Lest any of us get comfortable not meeting these stated goals, efforts are afoot to give the U.S. Office of Management and Budget greater funding oversight to reinvigorate the Federal Performance and Results Act of 1993 (see the recent news story in Nature [subscription required]).

Some argue that basic research should be held to different standards than applied research when it comes to performance review--you may have heard it said "of course the goals are rarely met, this is science!" The 1999 COSEPUP Report, Evaluating Federal Research Programs recognized the differences between basic and applied research, but still recommended that both types of research be described by performance plans.

While adages such as "take your worst estimate and double it" are often fairly accurate, they usually indicate less-than-ideally managed projects. Ksander admits that although you can study management techniques and be certified (see the Project Management Institute for examples) to the gills, nothing replaces real-world experience. On the other hand, experience without an understanding of management techniques also can be problematic. Future employers and granting agencies will be taking into consideration our ability to plan and execute a research program. It seems wise to seek out training in project management as we are gaining our independence.