Daniel Kahneman on improving the decision making process

IN MAY 2008 the McKinsey Quarterly spoke to Nobel laureate Daniel Kahneman, notable for his work on behavioural finance and hedonic psychology, about quality control and improving the decision making process.

1. The decision factory

Kahneman says that you can think of an organisation as a factory for producing decisions. The organisation might produce other things, but it produces decisions at all levels. Thinking about decisions as a product is a useful way to think about it, because it immediately raises the issue of quality control. As an organisation, whenever you have a product you take measures to ensure that your product meets certain standards.

2. Improving the quality of decisions

What can be done to improve the quality of decisions that are produced?

Kahneman indicates that quality control of decisions will be organised, in part, by bearing in mind those mistakes that are common and recurrent, and by making a deliberate effort to check whether those mistakes are happening.

Are there ways of eliciting the best information that is available in an organisation? Are the talents of the people that surround the decision maker utilised effectively? In many cases the answer is no. One could do a great deal better in utilising the human resources and information that are available in the process of decision making.

Unfortunately, there is a lot of hostility to the idea of improving decision making processes. Kahneman argues that as there is convergence to a decision, dissent becomes progressively more difficult and costly. Some changes, even though they may be desirable, are not adopted because they threaten the leadership of the organisation too much.  As such, people warning of potential disaster are increasingly set aside, shunned and treated as basically disloyal to the organisation that is committing itself to do something.

3. Pre-Mortem

Given that improving the decision making process is a difficult task, Kahneman presents the idea of the “Pre-Mortem”, an idea that he borrowed from one of his contemporaries, Gary Klein.

The Pre-Mortem is a simple idea, and involves a very straightforward procedure that benefits an organisation by helping to improve the decision making process. When you have a plan that is being formulated, convene your group for a short meeting. The meeting might be led by the person in charge or facilitated by someone else. Either way, the group should be presented with the following scenario: “it is a year from now. We have implemented the plan. It was total a disaster. You have a sheet of paper in front of you, write down a history of the disaster.” After giving everyone some time to write down their ideas you collect the pieces of paper and read them out. The process does not have to involve an extensive discussion.

The beauty of the Pre-Mortem is that it legitimises dissent. In fact, it does more than legitimise dissent. In organisations where the members are competitive, you expect people to think quite hard about the flaws in the idea and what could go wrong. In a room of twenty people you might expect three or four new ideas that can be used to readjust and improve the proposed plan of action.

4. Anchoring

Kahneman identifies anchoring as one of the big sources of mistakes in the process of decision making.

When you think about some quantity, like the amount of time it will take to finish a project, the first number that gets mentioned has an enormous impact on the way that people think.

Kahneman states that the psychology of anchoring is simple and happens automatically. There is nothing we can do about it. You retrieve a biased sample of information and then you evaluate that sample, and then it is already too late.

Kahneman provides a simple explanation of anchoring. Imagine you have a group of people. You ask them to write down the last four digits of their social security number, and then to consider a question, is the number of physicians in Manhattan higher or lower than the number you just wrote down? Now you ask the group, what is your best guess about the number of physicians in Manhattan? Kahneman suggests that what you are going to find is a difference of roughly 30% in the size of the estimates between the people with high social security numbers and the people with lower social security numbers.

When there is a particular number that is critical to a decision, try to trace down where the number came from, who brought it up first, and what information was used to support that number initially. Tracing down the history of a number is an example of something that is not too hard to do and, Kahneman believes, is almost guaranteed to improve things.

One implication of anchoring is that, if you are the leader and want to have an honest discussion about a number, you shouldn’t start with a number.

5. Opinions on paper, pre-discussion

Making discussions more fruitful and productive is something that we should think about. Kahneman provides one suggestion of how this might be done.

Before initiating a discussion solicit the opinions from each member of the group on paper. If you are going to generate an estimate of a number, for example what price is to be offered for a particular project, then getting opinions from the group on a slip of paper is going to improve the quality of the discussion.

This approach will help to draw out dissenting points of view. Once the discussion has begun, some group members may choose not to voice their dissenting opinion or may silently acquiesce to the prevailing view. Obtaining everyone’s starting position before the discussion ensues avoids this from happening.

In some cases, soliciting individual opinions before you discuss can make the discussion unnecessary. Kahneman gives an example of his involvement in an academic committee where he implemented this idea. In deciding whether to fund particular grants, the academics discovered that they loved to discuss, at great length, grants on which they all agreed. Soliciting individual opinions prior to discussion resulted in a significant reduction in the amount of time wasted on discussion.

🔴 Interested in consulting?