What Google's "Project Aristotle" Taught Us About Building the Perfect Team
Now that Tadaweb is in serious growth phase our team is expanding rapidly. This is an exciting time for many reasons and for me especially (the behavioural geek that I am), I find it extremely interesting to observe just how much the team dynamics are changing and evolving as the team size increases.... However, as the founder I believe that there is nothing more important than managing growth correctly, and a big part of that for me is ensuring we hire the right people, both from a knowledge but also cultural fit point of view. I am constantly striving to make our team the best it can be and really bring out the full potential of everyone on board. But then I got thinking.. Shit.. what's more important? Fostering potential at individual or rather a group level? How can we avoid the pot holes such as "group think" ? And well.. how is it actually possible to create a truly kick ass team as we grow?
I decided hit the books and researched journal articles until I randomly came across some incredible insight, quite by mistake, while reading Charles Duhigg's book ‘‘Smarter Faster Better: The Secrets of Productivity in Life and Business.’’ On finishing this awesome book, my own perceptions of how to build a successful team were completely disrupted. But why? Duhigg' uses the insights drawn from research untaken by arguably the biggest data geeks of all time - Google, in a project entitled "Aristotle" over 5 years ago in Google's quest to identify what made teams successful.
Ok, so let's take a deep dive into this cool bunch of work at Google.... Before the project Aristotle's researchers began analysing the behaviours of nearly 180 teams at Google, they did a review of over 50 years of academic studies that looked into team behaviour, nothing like a bit of exciting bed time reading. From that they were able to sculpt the key questions that would (hopefully) form the basis of their hypotheses. Questions included such things as if teams are more successful if the members share interests, socialise together out of work, share similar backgrounds etc. They asked if better teams included more extroverts or introverts? How did gender balance fit into the equation? How long had the team worked together? Then they measured a teams success..
However... in a very "un-Googlely" (it's a word, swear) result, the found.... NOTHING. Nope, zero, zilch, didly trends at all. No matter how they arranged and analysed the huge amounts of a data, not one pattern was to be found. No replicable matches could be made between personality types, backgrounds, diversity or anything. It started to become clear that the "who" aspect was not as important as one might think. This was a very eye brow raising moment for me! They noticed that some of the most successful teams at Google where comprised of friends who socialised a lot out of work, where other equally as successful teams were basically strangers, some had strong managers, others had a more flat approach to leadership. Some great teams were diverse, others as diverse as a ants nest. What was the most crazy however was that two teams with identical DNA and even overlapping members could have hugely contrasting levels of success! As a researcher, these type of moments are what you call a cluster f*ck.
In a scramble to make some sense out of the situation, the researches went back to the books again. Eventually they were onto something, that something was based around the concept of "group norms." You see, norms are like traditions, or standards of behaviour that are generally unwritten, yet still work to govern how the group works. Each team has a set of "group norms" that may illustrate why they prefer to debate issues out, or avoid disagreement. For me personally at Tadaweb, it is clear to see that each team have their own group norms, sys prefer to "debate" it out for all to hear, where as the web generally hit the meeting room at talk it over. Some teams arrive early before the buzz of the busy office is in full swing. Some teams eat together at least three times a week, others never do. Above all however, Project Aristotle found that group norms can have profound influence, going as far to over-ride how an individual performs in a group, compared to alone. Wow.
But.. the hard work was not over for the researchers. Group norms could help explain the ambiguity and apparent contradictions in the findings, but the true value comes from actually understanding exactly what group norms actually make a team successful. Google’s research had identified a bunch of of behaviours that appeared pretty important, however they weren't necessarily consistent across successful groups. To try and understand further, the researches then focused on a study conducted in 2009 by a group of psychologists from Carnegie Mellon, M.I.T. and Union College.
Now these guys weren't focused on exactly the same question as the Google researches, they were instead trying to work out was how intelligence manifests itself within a group - aka does a group composed of a whole bunch of super geeks get like "super duper geek smart" or does all their competing intelligence make them well.. dumber? Or rather, does a group of "average" intelligence bunch of people scale to get "smarter", a kind of "collective IQ." However, the results from their study proved hugely valuable to Project Aristotle! In testing their "group" hypotheses, the M.I.T. researchers ran experiments that asked groups to come up with innovative or creative uses for mundane objects, such as bricks. Some groups performed excellently, coming up with very unique ideas, others were, well shit at it, unable to create anything new. But what made the groups perform well, and what made the others do so bad? To answer their question researches found that when the right norms were apparent in a group, it was possible to increase the collective intelligence, where as bad norms could hurt it. However! That's not all they discovered, the most interesting result to emerge from the study was that the largest indicator of success in the group was in fact how team mates treated each other.
The researches, rejoicing in finally finding some "good search results" (see what I did there? Google, search results...? Never mind), continued on their side and begun to notice another couple of interesting behaviours that successful teams generally shared. First, members of successful groups generally spoke for the same proportion of time. Researches refer to this as ‘‘equality in distribution of conversational turn-taking’ a flash way of saying - they all took turns at talking. Conversely, they noted that if only one team member or a small group spoke the whole time, the collective intelligence actually declined! Shove that one in your mouth and smoke it Mr Chatterbox! The second behaviour of successful teams was to do with "social sensitivity" aka does Tommy know that Sally is sad by looking at her facial expression or hearing her tone of voice? (probably not because Tommy is a dude and dude's suck at that stuff **disclaimer, based only on personal experience and not scientific research).
If we want to get a little technical, within psychology, researchers often refer to traits like ‘‘conversational turn-taking’’ and ‘‘average social sensitivity’’ as parts of what’s known as psychological safety — a group culture that the Harvard Business School professor Amy Edmondson defines as a ‘‘shared belief held by members of a team that the team is safe for interpersonal risk-taking. Edmondson wrote in a study published in 1999 that "psychological safety is ‘‘a sense of confidence that the team will not embarrass, reject or punish someone for speaking up.’ Basically, the group needs to be characterised by interpersonal trust and mutual respect in which people are comfortable being themselves, yes even crazy cat ladies are encouraged to be crazy cat ladies at work... just minus the cats. There were a few other key components of successful teams, including the importance of goal making and creating a culture of dependability, however psychological safety took the ultimate cake.
Ok ok... I know this post is long.. (It's actually because I am on a loooong train ride from Brussels to Luxembourg, so you shall suffer too!). But what's most important here is that you can really just rock up to your team and say - "Jeff, can't you tell that Mark is f*cken mad?" Jenny - shut up, you are talking too much..." well, you probably could in the Tada office, but I don't think it would achieve true and lasting psychological safety creation. So how can founders use this concept to create kick ass teams in the real world?
I can't speak for everyone, but I can speak for my company and as the founder I try at least, to practise what I preach, with the exception of a few things when I prefer the quote "do as I say, not what I do." In our office we embrace sharing and one way we did this was on a team building day when I had each member of the team bring an item (tangible or intangible) that no one knew about to present to the group. Yes, basically a "show and tell" day for everyone to learn more about each other. One person explained their deeply personal tattoo that featured on his forearm, another woman shared her saxophone, another one brought a very special brick that he got from his old university on completing his MBA, another brought some home made juice. It seems crazy, but we created a safe place for each member of our growing team to get to know each other better through talking about their story. Above all, we respect family and as founders we always explain that at Tada, "family is first."
On a more day to day level, we talk a lot as a team, we share personal stories and we try and let people speak their minds in meetings. It is hard to turn introverts into extroverts, and in fact it's not a good idea, but through asking people their opinions it gives them the opportunity to express themselves. I can't speak for others' but I am pretty sure everyone feels safe here at Tada, people aren't scared to disagree or present their side of things. It has taken a long time for even me ot get my head around, but when I think about how many hours I work, it simple doesn't make sense to possess a "work me" and a "home me" to a certain extent at least. This has meant that I try and be myself as much as possible and having a supportive environment enables me to do so. Now... the next challenge is maintaining this as we get more people on board! Challenge accepted.