The Science Behind High Performing Teams

The Science Behind High Performing Teams

Two Critical Factors for Team Success

When you are selecting team members for a project, two factors are of critical importance in predicting how they will perform.  They are non-obvious and very powerful.  This article explains what they are and how you can use them next time you’re putting a team together to do something important.

Teams are responsible for most of humanity’s significant advances.  Sometimes through ‘standing on the shoulders of giants’, where innovation builds on previous discoveries. Other times, by the collective effort of a team of experts directed towards a specific goal.

Teams, whether separated by the mists of time or working contemporaneously, have invented the light bulb, discovered the structure of DNA and put a person on the moon.

Given the evidence you might reasonably conclude that teams are pretty smart. Reasonably but not necessarily correctly. Teams can be a hotbed of dysfunction and, on occasion, less than the sum of the parts. They don’t have to be.

Recent empirical studies have shed some light on how to create high preforming teams. Spoiler alert; the answer isn’t simply picking smart people, setting clear goals, and providing sufficient resource.  Theses may be important but they are not sufficient, as we shall see. But first a little background.



Collective intelligence has been a topic of interest for some time. In 1841 Charles Mackay published ‘Extraordinary popular Delusions and the Madness of Crowds’. One hundred and sixty four years later James Surowiecki published the ‘The Wisdom of Crowds’.  Two pretty opposing views that can’t both be right. In reality, though, they were looking at crowds in different ways; both of which are relevant in our context of business teams.

Mackay thought crowds exhibited ‘madness’ (his word, not mine) when they collectively believe in some bullshit theory or other.  That a tulip bulb could be worth more than a house (1623), that Internet start-ups could be worth millions of Dollars despite having no revenue (1999), that house prices would rise forever (2008) and that Brexit is a good idea (ongoing).

A micro version of these mutual delusions can infect business teams through the psychological phenomenon of ‘groupthink’. When a desire for harmony or conformity results in team isolation that leads to a lack of critical thinking.  Usually identified post hoc with the phrase “what the fuck were we thinking”. The cure for groupthink is pretty straightforward – introduce a formal contrarian role into the process. You can read more about this in my article ‘In Praise of the Devil’.


Surowiecki’s belief that crowds were wise was based on the observation that if you ask a large number of people to estimate an measurable quantity, like the weight of a prize bull at a state fair or the number of footballs packed into a Range Rover, then the average of all guesses will be very close the actual number.

This phenomenon is a very limited type of ‘wisdom’ but very useful for project teams.  I often run marketing workshops were attendees have to estimate key data for the purpose of developing a hypothesis – what would happen to sales if we increased price by 5%? How many more units would we sell if we offered an extended warranty? And so on.  The accuracy of the hypotheses, when subsequently tested, is often frighteningly precise.

When you follow a ‘Delphi’ process of individual estimates followed by team discussion followed by revised individual estimates you will not only often get a pretty accurate hypothesis, you will save a fortune in market research.

So teams/crowds/groups can be wise and they can also be stupid.  This much is self-evident.  The dysfunction, as it relates to groupthink, can be easily remedied and the wisdom, particularly around estimating quantities, can be harnessed effectively for very specific tasks. The question remains how do we select teams for optimal performance in the first place.


Thomas W Malone is a professor of management at the MIT Sloan School of Management and the founding director of the MIT Center for Collective Intelligence. When it comes to understanding how people work together he is pretty much the go to guy.  In his recent book, Superminds (2018), he poses the question “Can a group take an intelligence test?” His experiments in this area have some insight for anybody putting a team together.

We are all familiar with individual intelligence tests. They identify a single factor, such as an IQ score, that predicts with some accuracy how the individual will perform across a range of mental tasks.  Malone wanted to see if a single factor could predict group intelligence. For a full and interesting account of the process read his book. I can’t do it justice and getting data at source is always preferable than the second hand interpretation by someone looking to bathe in the intellectual afterglow.  Better from the horse’s mouth than the horse’s ass, as my mother is fond of saying.  So let’s jump straight to the results.

You might assume that factors driving team performance would include the intelligence of the individual members, how satisfied individuals are with the make up of the group, how motivated they are to achieve the goal and how comfortable they are working together.  Turns out none of these factors are significantly correlated with relative team performance.

What does seem to matter though is the social perceptiveness of the team members.  This means how well they can ‘read’ non-verbal clues to determine others feeling and emotions. When individual team members have a high score in social perceptiveness the team outperforms its peers. You can evaluate team members against social perceptiveness by using a ‘Reading the Mind in the Eyes’ test.  (Which I can provide if you can’t find a useable copy on the internet).

Other factors that had predictive ability included the proportion of women in the group or the tendency of one or two people to dominate discussions.  But when the study continued it became clear that the factor that had statistical significance was social perceptiveness.  The other factors may be important but possibly because there is an underlying mechanism at work that connects them to social perceptiveness.

One final factor that appears important is the mix of cognitive styles. Loosely speaking the three cognitive styles considered were verbal reasoning (good with words), visual reasoning (good with images) and spatial reasoning (good with components and structure).  In a business this might mean a sales person, a marketing person and an engineer or finance person.

Teams with an intermediate diversity of thinking styles performed better than those with to much similarity and those with too much difference.  Imagine a team made up entirely of sales people or accountants or marketers and you can see the problem. The answer, as in much of life is to be in the Goldilocks zone, avoiding the extremes.

This reflects what I frequently observe in my work. It’s the teams made up of sales, marketing, finance, operations, and HR, etc. which produce the better output than the more homogenous teams.

When you next need to put a team together make sure you have high levels of social perceptiveness in the team and a moderate and equal diversity of thinking styles. This gives you the best chance of forming an intelligent and high performing team.  Only then can you start worrying about clear goal setting and adequate resource.

Leave a reply

Your email address will not be published. Required fields are marked *