This article represents the confluence of three separate concepts I read about this week.
The first came when I read Michael Küsters’s article Why WSJF is Nonsense which details the downside of blindly ranking work packages based on the Weighted Shortest Job First (WSJF) formula. WSJF uses the ratio of the cost of delay to the relative effort required to complete a work package. It has become a popular method for prioritization as it avoids some of the problems associated with using a single metric such as business value. Michael’s concern is:
“We turn haphazard guesswork into a science, and think we’re making sound business decisions because we “have done the numbers”, when in reality, we are the victim of an error that is explicitly built into our process. We make entirely pointless prioritization decisions, thinking them to be economically sound. WSJF is merely a process to start a conversation about what we think should be priority, when our main problem is indecision.“
The second came from my re-reading of the chapter on the affect heuristic and availability cascades in Daniel Kahneman’s book, Thinking, Fast and Slow .
The affect heuristic is the idea that people make unconscious judgments based on their emotions related to the focus of the decision. If you view an advertisement touting multiple benefits of a product with which you’ve had no previous personal experience and are later asked about the risks associated with its use, you are likely to state that it is less risky even though there is no connection between its benefits and risks.
An availability cascade refers to the snowball effect which occurs when a relatively minor or infrequent event is blown out of proportion based on the emotional reactions experienced by those who are informed of it. The infamous Summer of the Shark in 2001 is a good example. Increased media hype occurred in spite of the fact that there were 76 shark attacks that year which was lower than the 85 attacks the previous year, and similarly, fewer shark attack-related deaths in 2001 compared to 2000.
Finally, I read a discussion thread on projectmanagement.com related to quantifying risk information. In the thread, Expected Monetary Value (EMV) was discussed and the proverbial lightbulb went off.
While EMV can be a useful way to quantify the expected impacts of the realization of a risk, it as susceptible to the same flaws which Michael and Daniel had written about. Unless there is a high degree of similarity in the contexts between the current project and past projects, quantification of a risk’s probability and impact is likely to be skewed first by normal estimation errors and further by our emotional responses to the risk itself.
If we have recently been on a project which was delayed because of the loss of a key team member, we are likely to give a much higher weighting to the probability and/or impact of a similar risk on our next project, even though these are entirely independent events.
Does this mean that we can’t use quantitative risk analysis tools? Of course not, but we should ensure we have some checks and balances when we use them to reduce the risk (pun absolutely intended!) of making a poor decision.
(If you liked this article, why not pick up my book Easy in Theory, Difficult in Practice which contains 100 other lessons on project leadership? It’s available on Amazon.com and on Amazon.ca as well as a number of other online book stores)