top of page

Intuitions - can we trust them?

Thank you for continuing to join me on this public exploration of decision-making with an aim to make better decisions.


This is the fourth blog in a series of blogs about decision-making. The previous blog (Jumping to conclusions (hartejsingh.com)|) covered how we can make predictions based on the data points we have seen which sometimes gives us misplaced confidence about our understanding of the situation, particularly if we only have limited experience.


This week I cover 1) what intuition is, 2) how it comes about and 3) when you should use it or not use it.


What is intuition? An intuition is arriving at a result, without conscious thought. We are surrounded by the language of intuition in literature, business, science and sport. Leaders who 'trusted their gut' are celebrated and some couples can draw their story back to the time they first met and 'knew instantly they would be together'.


The language of intuition


When we use our intuition we draw upon our rules of thumb and experience the idea that 'we've seen this movie before'. A popular book by Malcolm Gladwell called 'Blink' builds around the story of a world art expert looking at, what is meant to be, an ancient piece of art. Within a blink of an eye he felt that the piece looked too 'fresh'. Another character in the book was a marriage therapist who could tell within 15 minutes whether a couple would still be married in 15 years with a 90% accuracy. These are clearly well honed intuitions.


But intuitions are not always reliable, and the bat and ball problem has been proven time and again to catch people off guard.


The bat and ball problem


Give yourself 3 seconds to give an answer....



Particularly under time pressure, people often answer 10p (the answer is 5p). This is an intuitive answer and isn't miles off, but sometimes our intuition leads us astray with a lazy answer. More than 50% of students at Harvard, MIT and Princeton gave the intuitive (but wrong) answer. At least, with this problem, we can identify that intuition might sometimes catch us not fully engaged.


Intuition vs conscious thought? A fantastic explanation of this comes from the book 'Thinking Fast and Slow'. The mind has an intuitive system (System 1) and a thinking system (System 2). Consistent with the theme of the blogs so far, we take little energy to engage System 1 but take a lot of energy to engage System 2.


Humans tend to trust their intuitions and even try and defend an intuitive response with what seems like a rational defence after the event (post-rationalisation), but as we showed with the bat and ball test, people should not overly rely on their intuitions.



System 1 is your up-to-date model of the world and as we discussed last week, can sometimes jump to conclusions or give you a sense that you understand a situation with only flimsy data to back up your view. This will not be conscious but it will be comfortable.


Research shows that intuition performs badly at helping us consider things around numbers and extremes. Take this as a example. I recently asked a few people, roughly how much they would have at the end of one year, if they started with £1 and got 1% interest per day compounded. The answers that came back were nearer £5. The actual answer of £37.78. I am convinced it is worth avoiding intuitions around large numbers, rare events or probabilities. It is also worth avoiding generalising too much from small sample sizes (which I believe is the basis of much bias in this world).


So when is intuition useful? Let's take Malcolm Gladwell's marriage therapist as an example. His job has three elements of when intuition is useful:


1) Structure: there is a clear and measurable outcome (do they stay married?) with clear and measurable inputs (how they speak and act around one another)

2) Experience: It is someone who has seen the same or similar problem many times before

3) Feedback: They have kept score on whether their initial predictions were right or not and adjust their experience accordingly.


I would expect an expert chess player, fireman or school teacher to be able to make accurate predictions about their craft. BUT, and we all knew a but was coming, there are times when we might think we can rely on 'experts' but their craft does not have these three elements.


People might laugh at someone getting their palm read, but we regularly get presented with 'experts' in geopolitics, economics and stock tippers on the television who work in unstructured fields, and rarely get feedback on how their predictions pan out in the long run. In this way, we do not know if we can rely on their intuitions and would be well served considering this when weighting their advice. Whilst someone might always know their golf handicap (a meticulous measure of how many extra shots they need), for many other fields, keeping score does not exist.


So when should we use intuitions? Other than when we have a structured problem, with experience and feedback, we can justify using intuitions on decisions which are more subjective rather than objective for example, deciding on a colour of wallpaper. Secondly, where we lack time and an 'ok' decision is better than no decision, intuitions should be called upon. Finally, intuition helps when filtering between many choices, and can be used as a finishing touch to some analysis where there are many options (e.g. buying a house).


When and how do you use your intuitions?


How is this all relevant to decision-making? Here are three take-aways I want to leave you with before we pick it up next week:


1) An intuition is a result arrived at, without conscious thought. It's everywhere in our language but most commonly referred to as 'gut instinct'


2) We think we are better at intuitions than we actually are and can jump to wrong conclusions, but might reasonably be able to rely on them if we have a structured problem in which we have experience and a good objective track record


3) We should be careful when using the intuitions of others, particularly those that do not keep track records or in fields where it's tough to know if your prediction panned out or not


Next blog in a week's time will cover some of the textbook mistakes people make and how we can avoid it.


Link for further reading on this topic:


bottom of page