Science update: how to kill bad projects
Sometimes, it is comforting to see scientific confirmation for lessons learnt in practice. This happened for portfolio management in the March 2014 issue of the Journal of Product Innovation Management (JPIM), one of the most renown research journals in this domain. German researchers Behrens and Ernst found the best way to support “kill” decisions for bad projects (read it here):
1. visual decision aids, and
2. consultants advice.
They researched the question how to best prevent what is called “escalation of commitment” to a failing project. This phrase captures the tendency of managers to keep supporting a project they initially approved, even though it is turning into a bad project (destroying value). The research finds the best results for a combination of visual decision aids and consultant’s advice. Let’s have a look at both these angles.
Clear visualizations cause better go/stop decisions
The research work shows (building on previous research) that visualizing the course of a project contributes to better decisions. In this case, more managers actually stop a bad project with a clear view on costs and expected benefits. The visualizations are shown in the context of an initial expectation of value creation, so the project started as a “good” one. But over time it is slowly degrading by not meeting expectations. The best kill decisions are then taken before commercialization.
What about consultants?
When the researchers talk about consultant’s advice in their paper, they describe the consultant as the person independent from the decision maker, interpreting the visualizations and offering an advice. This may be an external specialist that is consulted for review. The same role can also be played by an internal independent analyst; the business controllers and portfolio analyst role come to mind.
What does it mean for the process?
In short, it means that the project and portfolio level decisions must be made in an environment with good and relevant visualizations, showing value and cost. Not just at one point in time, but dynamically updated over the projects, as the portfolio develops. Earlier research by Cooper already suggests that bubble plots contribute to better portfolios, this work builds on that by explaining why. So good solutions for portfolio management, such as our FLIGHTMAP platform, focus on decision support with the right set of visualizations.
In addition to the visualizations, the role of the analyst who comments on the meaning of the visualizations and can link recommendations to them, is essential. I am glad we have embedded this role in FLIGHTMAP’s decision workflow as well. Not because I have been this consultant for the better part of my working life, but because it leads to better decisions.