Submitted for publication by Email, February 19, 2009 © Robert Goatham & Calleam Consulting Ltd.
Published here April 2009.

Introduction | Projects From First Principles | Compounding the Challenge
Onion Rings | Experts and Expertise | Barriers to Developing Expertise | Conclusion

Projects From First Principles

Rather than attempting another round of quality improvement initiatives, perhaps the greatest hope for improving success rates lies in going back to first principles and understanding the very essence of what it is that makes managing IT projects so hard to do. Only then will we be able to develop practices and management approaches that truly overcome the problems.

The starting point for doing that lies in being able to step back from our traditional view of a project so that we can see the work we do at its most fundamental level. Traditionally, projects have been thought of as sets of discrete tasks that are linked to each other through dependencies. Most commonly, the tasks and their dependencies are represented using a Gantt chart. Each task is visibly delineated from the next and the relationships between tasks are clearly defined.

While a Gantt chart provides a useful visualization of a project, it represents a much-simplified view of reality. Although such simplifications are necessary, the danger is that it obscures the more complex reality. In the case of the IT industry, the gap between the simplified view and reality is significant and the space between the two is the breeding ground for the problems that lead to project failure.

Understanding the complex reality that technology team's face requires us to go beyond the relatively simple "task centric" view and look at projects in an entirely different way. At their most elementary level, all work carried out in a project can be thought of using four general categories:

  1. Physical tasks, such as laying foundations, building walls, etc.
  2. Information acquisition and analysis
  3. Information and knowledge transfer
  4. Decision making

The percentage of effort spent on each activity varies according to the type of project. While construction projects have a heavy bias towards physical activities, design type projects are dominated by the making of decisions. Figure 1 provides some representative examples of the different profiles of work that might be seen in different types of projects.

Figure 1: Typical percentage of effort by category of work
Figure 1: Typical percentage of effort by category of work

As Figure 1 shows, the work in an IT project is dominated by decision-making. Rather than being "physical tasks", the majority of the work we refer to as "tasks" in a technology project, is in fact "decision making". From strategy development all the way through to developing code, the primary activity that absorbs the majority of effort that goes into an IT project is the decision-making.

So, although the "task centric" view and the Gantt chart are firmly entrenched in our thinking about projects, rather than being a set of interdependent tasks, the reality is that IT projects are complex webs of interrelated decisions. This is illustrated in Figure 2.

Figure 2: Task versus decision centric views
Figure 2: Task versus decision centric views

So ubiquitous is decision making that we often don't think about our work in that way. While teams are generally aware of the key decisions made in a project, few will have considered that decision-making is the pervasive act that determines even the smallest detail of a project's outcome. The central role that decision making plays means that the level of success a project team achieves is directly correlated to the effectiveness of the team's collective decision-making capabilities. If the team is able to consistently make good decisions the chances are they will succeed. If the team makes too many bad decisions the chances of success are much reduced.

Introduction  Introduction

Home | Issacons | PM Glossary | Papers & Books | Max's Musings
Guest Articles | Contact Info | Search My Site | Site Map | Top of Page