Half the leadership in your organisation is not measured
Ask an organisation how it measures leadership and you'll get a clear answer. 360s on leaders. Leadership capability frameworks. Succession assessments. Development centres.
All of it aimed at individuals who hold leadership roles.
Now ask a different question. In a team that works well, who is actually exercising leadership? The honest answer is usually: more people than the org chart suggests. Someone spots a risk and raises it. Someone else notices a colleague is drowning and takes work off them. Someone reframes a stuck argument. None of them have "leader" in their title for that moment, and all of them are leading.
The research calls this shared leadership, and it turns out to be measurable.
What the meta-analysis found
In 2016, Lauren D'Innocenzo, John Mathieu and Michael Kukenberger published a meta-analysis in Journal of Management covering 3,198 teams.
Shared leadership was positively related to team performance. More importantly for anyone thinking about assessment, it explained unique variance in team performance over and above that of the vertical leader.
That phrase is the one to hold onto. It isn't saying shared leadership is a better alternative to having a capable leader. It's saying the leadership distributed through a team accounts for performance that the person at the top does not — that these are two distinct contributions, both real.
Two secondary findings with practical edges
The relationship was stronger where a team's work was more interdependent. That pattern has now appeared in several separate places in this literature: the more genuinely a team's members depend on each other, the more the team layer decides the result.
And how you measure shared leadership changes what you find. Network-based conceptions — mapping who actually influences whom — found more shared leadership than approaches that simply aggregate individual ratings. Measurement choice isn't neutral. Measure it crudely and you'll conclude there's less there than there is.
Why this matters for how organisations measure
If shared leadership contributes to performance independently of the vertical leader, then an assessment regime aimed entirely at individual leaders is systematically blind to part of what it's trying to measure.
That's not an argument against leader assessment, which the evidence supports. It's an argument that it's incomplete — and incomplete in a specific, identifiable way that could be fixed.
Fixing it is what the TeamHive 360 does. Shared Leadership is one of the four PLUS dimensions the instrument measures, precisely because this contribution is invisible to leader-focused assessment. In the University of Newcastle validation, the four team-level dimensions together explained variance in team effectiveness beyond two well-established individual-leadership instruments — which is this meta-analysis's point, observed in our own data.
Years ago I sat in on a team I had almost nothing to do with, having already decided its leader wasn't up to much. Too quiet, not who the organisation would have picked. I left having watched that team work better than any other I'd seen in the building, and wondering where on earth the leadership in that room was coming from. It was coming from everywhere, from everyone.
