You Cannot Kill Your Own Project
Handed $20 million and a failing project, the people who made the original call put $13 million behind it. Everyone else put in $9 million. The person best placed to see a failure is the one who can least afford to name it.
Barry Staw gave business students a company case and $20 million of R&D funding to allocate. Some of them had personally made an earlier funding decision that then went badly. Others had inherited someone else's failing bet.
The people who had made the original call put roughly $13 million behind it. Everyone else put in about $9 million. The worse their own decision had performed, the harder they backed it.
The driver is self-justification rather than stubbornness or bad arithmetic. Withdrawing funds is an admission, and the person best positioned to see a project failing is usually the person whose judgment is on trial. So they wait one more quarter for the number that vindicates them.
Which is why the review meeting so rarely kills anything. The room is largely the people who started it, and everyone present understands what the decision would say about the people in it.
The fix is structural: move the question. Have someone with no authorship review continuation, or write the kill criteria at kickoff, while nobody's reputation is attached to the answer yet. A decision made in advance costs nothing to honor later.
Source: Staw, Knee-deep in the Big Muddy: A study of escalating commitment to a chosen course of action, Organizational Behavior and Human Performance (1976)
