Ask most senior leaders what separates their best teams from their average ones and you will hear answers about talent — the right people, the right skills, the right experience. The research on high-performing teams consistently finds something different. The differentiating variable is not the quality of the initial composition. It is whether the team has built the structural conditions that enable it to keep improving over time. Superteams are not assembled. They are designed.
Ron Friedman's research on the highest-performing knowledge work teams, drawing on a survey of more than 6,000 individuals, identifies three traits that separate superteams from high performers who plateau. They get more done. Their members actively make one another better. And they keep improving. It is that third quality — the continuous upward trajectory — that is hardest to manufacture and most consequential for organizations running high-stakes strategic work over extended timeframes.
What superteams do that others don't
These are not personality characteristics. They are designed behaviors — structural conditions that leaders create deliberately and that teams practice with sufficient regularity to become self-reinforcing. The 50% higher experimentation rate does not emerge spontaneously from talented people. It emerges from a team climate in which trying things and failing instructively is treated as organizationally productive rather than individually risky. That climate is built by what leaders reward and what they penalize, consistently, over time.
The early obstacle surfacing — 43% more likely than average teams — is the behavioral output of a team that has learned its leader will respond to bad news with curiosity rather than blame. In the absence of that track record, the rational choice for any team member who spots an early warning signal is to wait — to see whether it resolves itself before escalating something that might reflect poorly on their judgment. By the time the signal is undeniable, the window for adaptive response has often closed.
What the strategy execution research adds
Wiita and Leonard's research on strategy-execution teams reinforces the superteam findings from a different direction. High-performing teams — those that successfully translate strategic intent into operational outcomes — spend significantly more time on two activities that average teams deprioritize. They spend 54% more time setting direction and crafting vision. And they spend nearly one-third more time building individual capability through development planning and deliberate investment in the team's skills over time.
The second finding is the one that most organizations systematically underinvest in. Capability building in high-performing teams is not a separate HR function or an annual development cycle. It is woven into the regular operation of the team — in how retrospectives are run, in what gets recognized, in whether the team's accumulated learning from past work is being deliberately transferred to the next challenge or simply forgotten.
The retrospective that most teams are running wrong
The standard team retrospective reviews performance against targets. Deadlines met or missed. Quality of deliverables. Stakeholder satisfaction scores. These are useful outputs but they measure what was achieved, not what was learned. A team that consistently hits its targets in the same ways, without ever experimenting with different approaches or improving its methods, is a team that has optimized for performance in its current context at the cost of adaptability in the next one.
The superteam retrospective asks a different set of questions. How many experiments did we run this cycle? What did we discover that we didn't know before? What did we try that didn't work — and what is the learning we are carrying forward? What obstacles did we surface early, and how did we use the lead time that created? These are not softer questions. They are strategically harder ones. And they produce the organizational learning that converts a team that performed well last year into a team that performs better next year.
The structural conditions that produce continuous improvement are not complicated. They require leaders who reward the attempt alongside the outcome, who make early honesty about obstacles feel safer than late honesty about failures, and who treat the team's learning as a strategic asset worth investing in deliberately. Teams that build these conditions keep improving. Teams that don't reach a ceiling and stay there.