Modeling risks with little experience seems like a crap shoot. But, modeling variables where we have experience can be equally dangerous due to familiarity bias. When familiarity bias creeps into risk modeling, it can be a serious problem.
What is Familiarity Bias?
A concise definition of familiarity bias is that we tend to underestimate risks in activities that are familiar. For example, driving is a dangerous activity, but most people would not think of it that way. People with a fear of flying, are more likely to experience a car accident driving to a destination rather than flying.
Another example is a project task that is relatively routine because we've completed it many times in the past and have data that shows how long it should take. To take it further, we've encountered unexpected difficulties in the task and overcome them. For the next project, we feel a strong understanding of the task and are comfortable with any contingencies. The problem is, what if the next contingency is something that's not easily handled? Just because we overcame past problems doesn't mean the next one is going to be just as easy.
If we're tasked with modeling the task time, it's tempting to go on past experience and overlook potential risks.
Avoiding Familiarity Bias in Risk Models
Understanding our potential bias is half the battle. Creating a model based on past experience (or previous models) should be treated with the same care as a new model. Drawing upon history and experience is still valuable and should not be ignored, but we still need to be creative and consider what we don't know despite the urge to go to the familiar.
Understanding potential familiarity bias in your risk modeling may save you from scratching your head as to why risk wasn't accounted for in a routine model.
Simulation Master is an Excel add-in for risk modeling. For more information please visit the Simulation Master product page.
Excel is a registered trademark of Microsoft Corporation. Used with permission from Microsoft.