Do you think you’re good at making decisions? Think again.
Whenever you are faced with a choice, any choice, your brain starts to work against you. Specifically, right after you take a look at all your options, you “fall in love” with a solution – often regardless of evidence and hard data. Once that happens, you want that solution to be right, and will oppose any other one, even more reasonable ones.
This is called a confirmation bias. In simple terms it means that your first instinct is to always go for a solution that you like, even if facts or data say otherwise. What’s more, even if you decide to take a look at all available evidence, you’ll still give more value to what supports your preferred decision and vice versa.
In practical terms, what does this mean to us?
It means that even when presented with all evidence you will not be able to be completely objective about your decision.
This, unfortunately, is reality – and it’s not just some people that suffer from it: it’s not a disease of sorts. The confirmation bias is a heavily-engrained mental process that is stuck deep in our brain.
The good news is that we are not completely powerless about it.
First of all, now you know about this bias which means that you’ll be able to recognise it when it appears. Second, even if we are not able to contrast it, we are very much free to go around it, and in this article I will share a few tools out of my book, Effective Decision Making, that will help you do exactly that.
Grid Analysis and KT Matrix
One way of going about it is to stop looking at the different choices as a whole and break them down in separate criteria. This way you will be able to address each criterion separately and remove the confirmation bias from your decision.
This sounds more complicated than it should be, so let’s take a look at two practical tools: the Grid Analysis and the Kepner-Tregoe Matrix.
The starting point is the same for both: you need a few alternative choices you could make – anywhere between two and ten is fine. More than 10 make the two tools cumbersome, although it’s still technically possibly to use them.
Once you have your alternatives, you need to list down what criteria are important to you, and that is the main point that allows you to reduce your confirmation bias.
Say, for example, that you need to buy a new security software for your company. You could be in love with one because most of your friends use it, but that is not necessarily the right one for you. Let’s assume you have three alternatives and your criteria for choosing one are the following:
The first step for a Grid Analysis is to plot all Options and Criteria in a Matrix.
Following that, you need to grade the performance of each separate option for each criterion with a number between 1 and 5 (or 1 and 10, whatever you prefer). Once you are done, add up each Option’s grades and compare the totals. The one with the highest score should be your preferred choice.
In our example, Option A is clearly the best option.
If you want to take it a step further, you can use a KT Matrix instead. This works similar to a Grid Analysis, but allows you to rank criteria as well. For example, Cost could be more important to you than Customer Service, so its weight should be higher in the analysis.
The first step is still to plot Options and Criteria on a Matrix, but instead of grading options right after, you should first rank how important each criterion is to you on a scale from 1 to 10.
Following that, just like for the Grid Analysis, rank each option on each criterion.
With a more refined analysis, our preferred option will be B.
Now, depending on the type of choice you are making you could go for one of these two tools or choose a completely different method (my book lists about 40).
Whatever method you choose, though, make sure it contributes to reducing the complexity of your choice. Sometimes we think that the more complex the method, the better – which is a very wrong perspective!
And as a last note, if you think brainstorming with your team could help, be careful! Brainstorming only works if some constraints are respected, and that is rarely the case (more on this here).
