I came across an interesting report about behavioural biases in decision making and project planning. This was extremely timely as I have been reading about biases and improving decision making lately, and this report ties together various principles with tangible impacts on project planning.
Link to the report: http://www.behaviouralinsights.co.uk/publications/a-review-of-optimism-bias-planning-fallacy-sunk-cost-bias-and-groupthink-in-project-delivery-and-organisational-decision-making/
I’ll do my best to summarize the key points, but I suggest reading it for yourself.
There are 3 common biases that impact project planning:
- Optimism Bias: the tendency for us to be over-optimistic when planning, underestimating the impacts of delays and overestimating our chances of success.
- Sunk Cost Bias: the tendency to take into account past investment in the process when making decisions, instead of looking forward at the remainder of the project.
- Groupthink: the negative impacts of social dynamics (pressure) on group collaboration/thinking
Cognitive biases and heuristics typically operate as ‘rules-of-thumb’ that allow us to make decisions quickly and efficiently based on past experiences. They form mental models in our minds (whether correct or not) of how to deal with a situation. So, how do these biases impact decision making and planning, and how can we overcome them?
We overestimate the impact of our own skills on past performances and underestimate the role of luck, known as the illusion of control. This is also impacted by our tendency to post-rationalize decisions, blaming unforeseen interruptions on “bad clients” and uncontrollable situations, standing by our decisions
We often take an inside-out approach, considering first the situation (and our personal attachment to it) and not to what has occurred in the past, known as the base rate. For example, when taking an inside-out approach when looking at a friend’s marriage, you would start by considering how happy they look and how well you know them, and predict that it’s unlikely they will ever get divorced. Taking the outside-in approach, you would start by looking at the standard divorce rate, and then make adjustments based on your knowledge of the situation.
A short-term solution is to build in an adjustment to all estimations, like the old heuristic of estimating development projects “Take your best estimate of what it will take and double it”. This may help, however there is a good chance that time spent and budget spent expand to fill the allotted amount.
Use pre-mortems as an attempt to understand potential failure points within a project, “In 6 months this project is completely off track and over budget. Why is that?”. By keeping track of past performances (time and budget performance) and referring to them at the outset of new projects, we can form better estimates.
In group situations, people will often change their thoughts/opinions in an attempt to conform to normal social expectations. This can also include authority or power figures unintentionally leading/guiding the discussion. Opening remarks made by one of these figures tend to guide the conversation by setting an anchor position that people rarely stray form, known as the cascade effect.
Group polarization is another common occurrence, as two differing viewpoints try to further differentiate themselves, pushing each towards the extremes. By controlling the order in which ideas are shared, such as not having the leader begin a discussion, you can allow a discussion to occur more organically.
- Use individual/small-group brainstorming before assembling to increase the diversity in thinking, encouraging people to come up with their own opinions ahead of time so they are less likely to be swayed.
- Consider having several groups plan a project completely independently to see how predictions differ.
Sunk Cost Bias
It is possible to be overly focused once a project starts and forget to come up for air. Sometimes your business/project landscape changes over time and you need to reassess as to whether you are on the right track and if it’s worth continuing.
Sunk costs tie into the optimism bias mentioned previously, causing people with sunk costs to become more optimistic with their projections. Status quo bias, related to sunk costs, is based on the fact that it’s far easier to maintain the status quo, than to change course. This is common in all things, such as trying to build a new habit or disrupt an industry.
Loss aversion, or the tendency to overweight losses compared to equivalent gains, compounds with the sunk cost bias to create desperate “all or nothing” situations where people are willing to take any gain whatsoever, compared to nominal incremental losses. For example, after spending $10,000 on a project, spending only $1,000 more for a potential $2,000 payoff seems worth it – numbers in this example are purely illustrative.
In order to overcome sunk costs, you need to focus on the present/future, ignoring what has happened in the past on this project. Ask yourself “If you were about to start this project today, would you continue?”
You will notice how quickly each of these biases tend to co-mingle and build off each other, leading to poor decision-making and predictions. Awareness of these biases is the first step to hopefully moving past them and avoiding them. Across all bias types, the role of “devil’s advocate” can help keep projects on check by intentionally pushing against consensus and forcing people to critically evaluate their position.
Using techniques such as making people to write down and submit their ideas/opinions in a brainstorm will help prevent people from being influenced by others’ thoughts, as well as force them to formulate an opinion on their own.
Overall the report was quite insightful, and will be useful when it comes to forecasting and project planning at Versett. I plan to (ideally) refer to these key points as we approach new projects, in an attempt to avoid some of these pitfalls.