In Waltzing With Bears, Lister and DeMarco describe the benefits of running IT projects within a portfolio. Not every one of them would have to succeed: you could take on several very high-risk (but high-payoff) projects and balance it with several low risk / low payoff projects. Having low-risk/high pay-off projects would be great, but most of the time those have already been done. You have a collection of projects that together will help you move towards your strategic objectives but do not have any single must-win projects where you have “bet the farm”.
The problem is that the idea can only be even really understood by someone who has gotten to Symbolic Parallel information processing (B4/ StrIV, for those who are keeping score at home). This corresponds to Director or Gen. Manager role with a TSD of 2-5 years. Once again, managers cannot delegate planning to subordinates because they will not have time horizons sufficient to allow them to see the whole picture. If they do, they’re probably going to be gunning for their boss’s job.
I wonder if it isn’t a level above that. The projects are interrelated, sharing resources and combining to form a single goal. That seems to imply a meta concept that is bigger than the interrelatedness of the projects themselves. Which would put this idea up even farther. Anyone got any comments on that?
I wonder if Lister and DeMarco could have come up with this idea when they wrote PeopleWare.
Image Credit: Mountains near Château-d’Oex. © E. Forrest Christian. All rights reserved.