which the project was initiated. Their point—that we need to be thoughtful about how we measure project success—is well taken. However, I think the logic of their example is flawed and subject to dangerous misinterpretation.
I would assert that the only people who can determine whether or not a project is successful are the project’s sponsors—the leaders (usually executives) in an organization who have responsibility during the project for deciding whether or not to continue investing in it. If you were on the Sydney city council, you might have imagined that the Opera House was worth an expenditure of $3-$4 per person for your citizenry in 1957 (several hours’ wages at the time). When the cost ballooned to $40-$50 per citizen (a week’s wages for many) and you faced difficult choices about cutting spending elsewhere, you might have felt differently. The sponsors get to decide whether the project is worth the investment and the risk. The sponsors get to decide whether to maintain, expand, or curtail scope to meet budget or schedule objectives. The primary role of the project manager is to support informed sponsor decision making.
I first read the book in 2008, but an NPR story about several cities in the US that are literally on the verge of bankruptcy due to runaway public works projects renewed my interest in the subject. One of the examples in the NPR story was Harrisburg, PA’s incinerator project that initially cost $15 million but had a final price tag of $288 million. My intention is not to kick the nice folks of Harrisburg when they are down or to fault the project manager of the incinerator project (I don’t know the details), but instead to offer their story as a cautionary tale to any project managers who might be tempted to lose sight of whose interest you are expected to serve.
Projects (particularly IT projects) can be difficult to estimate. Sometimes this is due to poor planning, sometimes it may be a consequence of unforeseen circumstances, and sometimes unknowns have a bigger impact on our projects than we expect. All a good project manager can do is be rigorous and try to inform sponsors about the likely time and budget requirements as well as the likelihood that those requirements are reasonably accurate. Sometimes surprises arise in the middle of a project. When a project has begun and actual cost and schedule performance data becomes available, project managers should provide the data should to sponsors to inform decisions about whether and how to continue.
I believe successful project management is about supporting informed sponsor decisions. Did the project manager provide timely and accurate information? Were decisions made at an appropriate organizational level? Were the right decisions escalated? Was risk managed consciously? Successful project management may not always lead to a successful project. Sponsors may make informed decisions to accept risks that undermine a project’s business case if they materialize. This is not a failure of project management.
I would argue that project success can only be determined by the project’s sponsors—not the team, observers, pundits, or consumers of the product the project creates. If sponsor goals are achieved within acceptable boundaries of time and resources, then sponsors should consider the project successful. Sponsors are allowed to modify or redefine their goals at any time during the project. In my view, determining success is not the jurisdiction of spectators after the fact, based upon how much they appreciate the project results. If that were true, the Harrisburg incinerator might be considered a successful project.
It seems to me that if we want projects