Ten reasons teams really matter (and why I had to jump through hoops to see it!)

Over the past few weeks we've published eight "Teams Matter" articles on the TeamHive blog, each one working through a piece of research on teams. Google's Project Aristotle. A study in Science on collective intelligence. A meta-analysis of over three thousand teams on shared leadership. McKinsey on top teams. Six thousand papers mapped by a group at the University of Twente.

I wanted to pull it all into one place, partly for the people who've asked me "so what's the short version?", and partly for me. Because reading the whole series back to back, what struck me wasn't the research. It was how long it took me to believe what the research had been saying the whole time.

So here's my list. It isn't really a top ten. It's more the ten things I hold onto now when someone asks why I've spent the last few years on teams rather than on leaders. Some come from the studies, some from rooms I've sat in. Most from both.

1. The leadership in a great team is everywhere, from everyone.

I've told this story before and I'll keep telling it. Years ago I sat in on a team I had almost nothing to do with (I honestly can't remember why I was in the room), having already decided its leader wasn't up to much. Too quiet. Not who the organisation would normally put in charge. I left humbled. That team hummed like none other I'd seen in the building, and when I asked myself where the leadership was coming from, the answer was everywhere. From everyone.

It turns out that isn't just a nice story. A 2016 meta-analysis of 3,198 teams found that shared leadership predicted team performance over and above the contribution of the formal leader. Two separate contributions, both real. And yet almost every organisation I've worked with measures only one of them. Leader 360s, capability frameworks, succession assessments, all aimed at the person with the title. Half the leadership in most organisations simply isn't measured.

2. Brilliant people don't add up to a brilliant team.

I've sat in rooms full of genuinely clever people and watched them produce mediocre work. I've also sat with teams that looked unremarkable on paper and watched them run rings around the clever room. For years I filed that under "chemistry", which is another way of saying I didn't have the language for it.

The language exists. In 2010 Anita Woolley and her colleagues put 192 small groups through a battery of very different tasks and found that groups have a general intelligence of their own, a collective intelligence. It barely moved with the average intelligence of the members, or with the smartest person in the room. What it tracked was behaviour. Groups where people took roughly equal turns talking did better. Groups who were better at reading each other did better. The capability wasn't sitting inside the cleverest member. It was in the interaction.

3. Even Google couldn't make composition win.

This is the study I bring up when someone tells me team quality is really a hiring problem. Google's People Analytics group spent two years and 180 teams trying to find the mix of individuals that explained why some of their teams thrived and others struggled. They're Google. If anyone was going to find it in the data, they were. They didn't. What mattered was less who was on the team and more how the team worked together. Team size, seniority, whether people sat in the same office: no meaningful relationship.

I'd spent more than a decade seeing exactly that pattern, without their data to check it against. Talented people working alongside each other rarely, magically, produced a better team.

4. Individual development on its own was hit and miss. I know, because I did it for ten years.

This is the one I got wrong for the longest. For over a decade my main approach to culture and team performance was leadership development. Send the leaders on a program, coach them, build their skills. It produced nice positive impacts. Whether real, sustained transformation followed, whether their team actually became more than the sum of its parts, was honestly hit and miss.

It was only when we shifted to sustained team coaching at Brave Insights that we saw top teams consistently change, and the culture, the engagement scores and the business results move with them. We tested this properly in a pilot at an Australian financial services company, alongside an organisation-wide culture program, measuring behaviour and capability change over time. Every intervention helped. Workshops, peer coaching, individual coaching. But team coaching showed by far the biggest shift in long-term behaviour, for individuals AND for teams. That result is a big part of why TeamHive exists.

5. How the team works is the lever you can actually pull.

Most leaders I work with are a bit stuck on composition. You inherit a team. Hiring is slow and constrained. Moving people is disruptive and often politically impossible. If team quality came down to who's in the room, most leaders would have very little room to move.

In 2015 the US National Research Council reviewed decades of research and concluded, without hedging, that team processes are the primary point of influence on team effectiveness. Not composition. How the team actually works. The relationship was more than twice as strong where the work was highly interdependent, which describes pretty much every leadership team I've ever sat with. Process is the one thing a leader can start changing on Monday morning without a requisition or a restructure.

6. It's the boring stuff that predicts performance.

Not the conference-title ideas. The mechanics. How a team plans between bursts of work, how it coordinates and backs each other up while the work is happening, how it handles conflict and keeps itself workable. A 2008 meta-analysis by Jeffery LePine and colleagues found these everyday teamwork processes predicted both performance and how satisfied members were, and the pattern held however they sliced it. Findings that survive being sliced several ways tend to be findings about the world rather than about the questionnaire.

I find this oddly reassuring. The things that make teams work are ordinary and learnable. Taking turns. Reading the room. Agreeing who does what. None of it requires better people than the ones you already have.

7. It still holds when performance is a count of mistakes.

The objection I hear most often, usually from someone with their arms folded, is that team research is all surveys. People rate their teamwork, then rate their performance, and of course the two move together.

It's a fair objection, and it has an answer. In 2019 Jan Schmutz and colleagues pooled 31 studies covering 1,390 teams in emergency departments, operating theatres and intensive care, where performance is an error or an expert's rating recorded by someone outside the team. Teams with better teamwork were around 2.8 times more likely to perform well. And what was being measured on the input side was behaviour: coordination, communication, anticipating what a colleague needs. Not personality. That's a design decision for us at TeamHive too. We measure observable behaviour because that's where the signal lives.

8. Your top team is a financial performance variable, and CEOs already know it.

Everyone agrees teams matter. Almost nobody puts them in the same conversation as capital allocation. McKinsey's work with nearly 200 CEOs found that companies whose top team is aligned and working well together are almost twice as likely to deliver above-median financial performance, and the large majority of those CEOs named building an effective top team as a priority.

When I read their description of what those teams actually do (shared direction, constructive challenge, shared accountability, the capacity to adapt) I recognised it straight away. It's the territory of Purpose, Unity, Shared Leadership and Learning. We built PLUS from practice and had it validated at the University of Newcastle. McKinsey got there from two hundred CEOs. The maps match, and neither of us copied the other.

9. This is one of the best-evidenced fields you will ever be asked to fund.

"We'd love to invest in teams, but the evidence isn't there like it is for other things." I've heard some version of that more times than I can count, and I understand why. Team development often arrives wrapped in language that invites scepticism.

But as a description of the research, it's just wrong. A bibliometric review published this year mapped 6,051 peer-reviewed publications on team effectiveness over thirty years, growing exponentially, organised into ten established sub-fields. That's a mature discipline. Treating team effectiveness as the unproven end of organisational development is a choice. It isn't a reading of the evidence.

One small thing in that review made me smile. They define team effectiveness as achieving performance objectives, meeting stakeholder expectations, and sustaining the team over the long term. Two of those three are invisible from inside the team. It's why the TeamHive 360 asks the people around the team, not just the team itself. Something I'd insisted on for years mostly on instinct.

10. Once a team sees its reality, it moves. Fast.

This is the thread that ties my own journey to all of the above. In all our team coaching, the biggest progress came once a team had got to its core issue, the heart of what needed to change. Once they'd confronted reality and had the insight, the change itself happened at a much greater pace. Which made me wonder: if teams were armed with the right information about their reality, could they have that insight and move themselves, with less external help per team?

That question became a research partnership with the University of Newcastle's School of Psychological Sciences, hundreds of survey items, an expert panel, and a validation study with 500 Australian employees each from a different team. When the report came back I was blown away. The four PLUS dimensions together explained 60% of the variance in team effectiveness. We could measure the thing I'd been watching in rooms for twenty years.