Portfolio Management

Any organization with a variety of significant projects is eventually forced to do portfolio management, that is manage the "investment portfolio" that those projects represent. Like any investment package, any given project may rise or fall in value depending on the total mix and on the fluctuating environment. This has the frustrating consequence that even well managed projects may be cancelled to enhance the overall value of the total portfolio.

Part of portfolio management is an optimization problem. Projects have associated values and probabilities of delivery. Some sophisticated calculations can generate an expected net return on the portfolio. Applying the same sequence of calculations repeatedly to different scenarios would allow portfolio managers to find the advantages (or disadvantages) of any proposed change to the portfolio.

The life or death of individual projects can also be the or death of individual careers, or at least the individual motivations of staff. While an optimization calculation may mandate the cancellation of a project, that action could produce a ripple effect that goes well beyond the intent of the portfolio managers. Key engineers could quit. Other project managers could become defensive or anxious. In one of my clients, an engineer on a cancelled project left and took important technological knowledge and client contacts with him to a competitor.

The message here is that a portfolio is not a "bag of marbles" (i.e., independent objects); it is a collection of objects with rich connections to a myriad of other objects in the organization: people, careers, feelings, as well as other projects.

A portfolio of projects is also a way to hedge our bets about the future. Sometimes a portfolio may contain projects based on competing technologies. The assumption is that at least one of the projects will be appropriate for some unpredictable future time.

Projects may also reflect differing assumptions about competitor behavior, or regulatory trends. The net result is that portfolios are often internally inconsistent, precisely because they anticipate conflicting future possibilities.