Eliminate Errors in Decisions

Three strategies to reduce random and unpredictable errors in human judgement, highlighted from the book 'Noise' by Daniel Kahneman.

Last Monday, my cat Bubbles gave birth to a new litter of three kittens. I spent the day with her as she went through labor. Here's a picture of the new babies with Bubbles and Snowy.

My latest non-fiction read has been Noise by Daniel Kahneman and Cass Sunstein. The book is about errors in human judgment, specifically, the random and unpredictable errors that are more insidious than easily detectable ones caused by bias. These random and unpredictable errors are called noise. The authors assert that wherever there is human judgment, there is noise and more of it than one would expect.

The book is a fascinating read, and I highly recommend it, primarily if you work in fields that require human decision-making. The authors also outline a few key strategies we can adopt to reduce this noise in our decisions.

This edition focuses on the three most critical noise-reduction steps that you can take to reduce errors in your decision-making.

1️⃣ Mediating Assessments

The first way to reduce noise in decision-making is to break the decision into smaller, separate questions. "Should I hire her?" might be a big decision that requires a complex evaluation that might be skewed based on first impressions or your personal priorities of how you evaluate a person. Or, it might be influenced by irrelevant events that have nothing to do with this decision- maybe you're in a terrible mood because of a fight at home, and you're uncharacteristically harsh in evaluating candidates today.

But if you break a big question down into smaller, separate questions that can be evaluated independently, it gives you unique insights into diverse aspects of the decision. For instance, you could outline a set of questions before the interview: what is the candidate's educational background? What are the highlights of the recommendations they bring from their previous roles? What relevant experiences do they bring to this role? Then, during the interview, you will actively look for this information instead of taking it in a direction based solely on your first impressions or factors that might be irrelevant for the job you're hiring for.

When you break a decision into a series of smaller, separate questions that can be answered independently, you engage in a mediating assessment protocol. Mediating assessments help us to avoid biases and snap judgments that typically cause errors in decision-making.

2️⃣ Delay Intuition

There is value in letting people make decisions based on personal intuition or 'gut-feeling- indeed, everyone craves an opportunity to express themselves through judgments and love the power to make decisions. It is why people want promotions and leadership roles, and it is also why they will lash out if you try to curtail this ability.

However, the purpose of decision-making is accuracy and not human expression. As long as we can agree on this, we will also need to increase accuracy in judgments while reducing errors. The most important step we can take is to delay human intuition in the decision-making process. It means that while we don't ban the human element of decision-making, we take all steps to ensure that it is the last step. This gives us a chance to take in more information, look at the case from several angles, discuss with experts, and more generally, make better decisions. Then, even if we believe we used our intuition to come to a conclusion, at least this 'intuition' will be more informed, educated, and more accurate than snap-judgments that are prone to errors.

The mediating assessments protocol I outlined above is one of the tools that help to remove errors of biases and "gut-feeling" induced judgments. Another tool is to avoid irrelevant information from reaching decision-makers. The book lists an example of fingerprint examiners whose judgments of whether the crime-scene prints matched with the accused were heavily influenced by details of what other fingerprints examiners decided. So if you knew that your colleagues judged that the prints matched, you were disproportionately (and unfairly) likely to agree with them. This was a subconscious influence since the examiner gave a different answer and explanations on their reports when they had no information on what other examiners had decided.

When we make important decisions in our own lives, we should limit our investigation and the information received to strictly relevant factors. For instance, if you're hiring someone for a job and believe that personal opinions and lifestyles do not influence the work performed, limit the information strictly to relevant factors and references. Stick to their LinkedIn profile and professional references, and avoid their personal social media profiles on Instagram or Facebook where they're likely to display a side that is not relevant to work. Again, stick to essential and directly relevant factors, and avoid irrelevant information that might influence your emotions or immediate reactions, and therefore your decision.

3️⃣ Rank Your Choices

The human brain finds it easy to compare two things. When you're given two movies, it's easier for you to decide which one you like better and explain why. But the brain finds it more cognitively demanding to evaluate a choice in isolation or absolute terms.

It is actually easy to understand why: making a decision triggers a complex neurological process to evaluate a host of questions, decide pros and cons, create rules to make exceptions for that particular option. So, when we assess a choice in isolation, we need to start this process all over again for each choice. On the other hand, when we're asked to compare between choices, our options are bounded, and we only need to decide how to rank these options. So we can easily explain why we ranked three candidates for a job in a particular order. But we find it harder to provide accurate reasoning behind why one specific candidate qualifies for the job or not.

So, the last strategy to improve your decisions is to make relative judgments rather than absolute ones. Our brains are better at making comparisons.

Objective, isolated decisions are error-prone because they are disproportionately cognitively demanding. Further, every instance of this process is easily affected by irrelevant factors such as your current emotional state, physical discomfort, what time of the day it is, or if your favorite team just lost their match.

Relative decisions are less error-prone because they are cognitively easier, and you usually only have one way to arrange a fixed number of choices.

You could implement this in two ways. The simple way is to arrange all your choices in a ranked order. This is easy when you already have all your options and have to decide from them. Instead of deciding on each of them separately, assign all of the ranks in comparison with each other. You could give them all a score to be more objective about it, but your decision essentially boils down to: "Is this option better or worse than the previous one?".

The second way is to create a 'Case Scale'. Using your previous experiences of making similar judgments, create a scale placing known decisions at different places as anchors. For instance, if you previously decided to hire an excellent candidate who went on to be great at their job, place them as a case anchor on one side of your scale. On the other end, you could put a candidate you rejected due to lack of experience or other relevant reasons. When you have several such anchors on a scale, every subsequent decision can be compared against an anchor to check where this new candidate would be placed on this scale. In this way, we turn an isolated decision into one based on a comparison.

These have been my key takeaways from Noise by Daniel Kahneman and Cass Sunstein. The strategies outlined in this book are especially instructive for those working in areas that require frequent decision-making. I'd love to hear from you on how you might apply some of these ideas in your life.


These have been my three highlights from the productivity and personal growth space this week. If you liked what you read, connect with me on Twitter and let me know. If someone forwarded this newsletter to you, subscribe to GrowPro Labs here to receive it in your inbox every Monday.

I'm on a mission to make Mondays great, and this newsletter is a part of my efforts in helping you achieve mindful productivity, personal growth, and living your Ikigai.

-Vensy