Skip to main content

Learning from Detroit, or, Why projects fail

When I heard about the would-be Christmas bomber, I couldn’t help thinking, “Why haven’t the 9/11 Commission Report recommendations been implemented?” And since that day, indeed, we’ve come to find out that the various security agencies are still not talking adequately to each other.

This has brought to my mind something I’d read about a year ago in the PMI’s Project Management Journal: System Biases and Culture in Project Failure (

It turns out that, when we make decisions, we are influenced by nine distinct “systemic biases:”

  • Available data: Preferring easy-to-get information over anything else.
  • Conservatism: Rejecting new or contrary information.
  • Escalation of commitment: Increasing the resource investment, never mind those flashing red lights.
  • Groupthink: Adopting a party line–and sticking to it.
  • Illusion of control: Believing in our own power over things we can’t control.
  • Overconfidence: Feeling better about the state of things than the state of things can support.
  • Recency: Thinking fresh information is best.
  • Selective perception: Lacking a common understanding of the same set of circumstances.
  • Sunk cost: Throwing good money after bad.

Barry Shore (, the author, describes seven major failures, some of them famous (like the Denver airport baggage handling fiasco) to build a careful argument as to “how the rational processes of project management can be derailed by the human decision-making process.” Interestingly, the organizations involved all had something in common. As he writes, “failed projects, in general, can be associated with organizational and project cultures characterized by an internal focus and a preference for stability, not change.” (I’ve added the emphasis.)

And, it’s almost a truism that the more complex the project and its players, the easier it is to hide (and hide behind) these behaviors. I’m thinking that these ideas are worth keeping in mind as we head into a year of major initiatives, ones that cross organizational boundaries and challenge the way we have done business in the past.

I’m also wondering what the chances are of shining a light on our decision-making processes so that harder-to-get and legacy information make it into the flow, contrary voices are heard, exit strategies are built and allowed to be taken, realism and a sense of proportion rule, and the project participants (however geographically spread) share a common understanding of scope, status, and roles.

Phew! That’s probably too high a bar if we set it as an absolute. Clearly, though, project management at the scale where we now need to work has to go hand in hand with cultural transformation. This is one message I’ve taken from the events above Detroit on 12/25/09.