What makes a team perform well? Most studies tend to rely on anecdotal evidence to determine the markers of success. But perhaps one of the best studies on the subject is by MIT’s Alex Pentland. He outlined the conclusions in his 2014 book: “Social Physics: How Good Ideas Spread – The Lessons From a New Science”. He has essentially accumulated vast amount of data on the social interactions of teams and companies by attaching devices (like sociometric badges) to individuals.
His main finding was that top-performing teams were those where each person spoke for an equal amount of time and were good at reading each other’s social cues. What did not matter was the individual intelligence of each team member. It was the quality of interactions, not the quality of the individuals that mattered. This turns the focus of companies on hiring and retaining stars on its head.
In addition to the above, here were three other take-aways:
- Open people, better networks. The most insightful people seek out people with different views and different ideas. They then winnow down their most recently discovered ideas to the best ones through their habit of bouncing them off everyone they meet. The more diverse the network the better. Not only that, but stars look beyond just doing their part of the team’s output, they push the team jointly with their own goals, and make everyone feel part of something more than their job. It follows, then, that the most successful leadership style is the charismatic connector. They are genuinely interested in everyone and everything. Ask about what is happening in people’s lives, how projects are doing, how they are addressing problems. Finally, the more direct interactions between people, the more trust will build.
- Learning the social way: The optimal way of learning is to spend 90% of your time finding and copying others who appear to be doing well, and only 10% on individual experimentation and thinking. Not only that, but our desires and preferences are based on what our peer community agrees is valuable, not on our individual rational reflection or in-born morals. Consequently, it has been shown that multiple exposures that a new behaviour has a good outcome are needed for habit to change. Who vouches for or is associated with a message is four times more influential than the message itself.
- Incentivise interactions: Provide incentives aimed at social networks, not at changing individual behaviour. This has been shown to be four times more effective than individual incentives. The biggest targets should be on people with the strongest social ties and most social interaction with others. Also, by adding a social component to performance, for example a comparison to peer groups (think tax returns completed in your neighbourhood), it can be twice as effective than just using an individual incentive alone. Providing visual feedback of interaction patterns also helps. This means away with organisational charts and in with interaction maps.
There is a dark side to harnessing the social capital of an organisation. Most notably herding, where social influence overwhelms diversity of views. This needs to be countered by a keen eye on the origin of ideas, discounting common opinions and keeping track of contrarian ideas.
The bottom line, which I agree with, is that Alex views successful organisations as those that promote healthy interactions, not individuals or specific content. The numbers back this up, as he finds that healthy interactions accounts for half of the performance variation between high and low performing groups. Time to change the focus of companies!