It is useful to know about the queuing theory. Especially, when dealing with project portfolios, understanding the Little’s law is important. The law states that
The long-term average number of customers in a stable system L is equal to the long-term average arrival rate, λ, multiplied by the long-term average time a customer spends in the system, W; or expressed algebraically: L = λW.
Applying this to the project portfolio management is rather straightforward. The number of projects in the portfolio is the product of establishment rate of new projects and the average length of projects. Thus, if there are too many projects, one can either start fewer in the future or speed up the existing and future ones.
But how can one speed up existing projects? If projects tend to take too long, there are two ways for speeding them up as W = L/λ. However, it is not easy to change the arrival rate of new projects, as it depends on business needs and it not something the ICT department can readily affect. Thus, the easiest way to speed up projects in the portfolio is to cancel some of the existing ones. But which ones should be cancelled first?
Fortunately, there is a simple answer. One should cancel all IT projects, or projects which provide no immediate/proven business benefits. The easiest way of spotting IT projects is to check which projects do not have real sponsorship from business: no testing support, no working steering group, no commitment whatsoever, no project sponsor. Any such project should be cancelled. Another class of unnecessary and thus usually harmful IT projects consist of projects that simply replace an existing system with another without changing business processes – these projects hardly ever produce any lasting value.
By pruning IT projects from the project portfolio the ICT department can serve the needs of the business much better.