Harvard Business Review's May/June 2026 issue carries a single headline argument that most L&D and leadership budgets are not currently designed for. In How to Build a Superteam That Keeps Getting Better, the cover article by Ron Friedman, the finding from a study of more than 6,000 knowledge workers across finance, law, healthcare and technology is blunt: "In periods of rapid change, the teams that outperform everyone else are not those with the best plans or the most talent but those that learn the fastest."
That sentence has a procurement implication most people read past on the first time through.
What The Study Actually Found
Friedman's data names three strengths that distinguish higher-performing teams from the rest. They run their time, energy and attention more efficiently. Their members actively make one another better. And they keep building skills, in small increments, over the lifetime of the team rather than in the week of a programme launch.
The leadership behaviours that produce those strengths are equally specific. Friedman names seven: running more experiments even when things are going well; making intellectual humility and curiosity contagious; asking the unglamorous question "what are you stuck on?"; rolling up sleeves and staying close to the work; making feedback feel like support rather than verdict; investing in growth even when the payoff is not obvious; and leading with meaning rather than only metrics.
None of those is a slogan. Each is a behaviour a manager either does in a particular Tuesday meeting, or does not.
Why The Standard Diagnosis Misses
The default response to a team that is not learning fast enough is procedural. Add a quarterly off-site. Buy a learning platform. Run a workshop on growth mindset. None of those touch the moment a senior person, in front of a junior one, has to say "I do not know yet, what would you try?" and mean it.
Two findings from the wider research literature explain why awareness training so rarely shifts learning velocity. The first is Roediger and Karpicke (2006), in Psychological Science: being asked to retrieve and apply material increases long-term retention by around 50% compared with re-reading it. A two-hour module on psychological safety behaves like the re-reading condition. Six weeks later, the manager who is meant to model intellectual humility is back to defending the plan.
The second is Edmondson (1999), in Administrative Science Quarterly, on psychological safety, defined as "a shared belief that the team is safe for interpersonal risk-taking." The behaviours Friedman is naming, asking what someone is stuck on, surfacing problems early, treating feedback as support, all live downstream of that one belief. Google's Project Aristotle, the multi-year study of team effectiveness, ranked psychological safety first among the dynamics that distinguished higher-performing teams. Different research tradition, same construct.
What Most Organisations Do (And Why It Stays Slow)
The recognisable 2026 pattern is a leadership programme front-loaded with content, a quarterly review heavy on metrics, and a retrospective ritual that lasts twenty minutes. Curiosity is praised on the values poster and quietly punished in the meeting where it slows the agenda. Experiments are encouraged in the abstract and discouraged in the budget cycle. The team learns at the speed of the calendar, not at the speed of the work.
Friedman's data does not support reading that as a "talent problem". The same workforce, under managers who were equipped to model the seven behaviours, moves measurably. Slow learning, in the data, is mostly absence of rehearsal at the layer above.
What Works
The teams whose learning velocity, retention and output have moved upward share three habits.
They define the behaviour, not the value. The point is not whether managers say they value curiosity. The point is whether, in the meeting where the deadline is tight and an analyst raises an awkward question, the manager pauses, says "good catch, what changes?" and keeps going. That is a skill, and skills decay without rehearsal.
They put the difficult moment in front of managers in a low-stakes setting before the real one arrives. Building on academic behaviour-change work from UCL, Cambridge and Bocconi, our own research found that immersive role-play with professional actors was around 20% more effective than passive modalities such as slide-show or video e-learning at moving observed skill. The same study found participants overestimated their skill before measurement, the Dunning-Kruger pattern that quiet meetings reward.
They re-measure. Not training completion, not engagement-survey averages. The honest test is whether the average team retrospective in month nine surfaces three problems that were not raised in month three. If it does not, learning velocity is a slogan, not a property of the team.
This is the work behind our behaviour-change labs and our immersive simulations. Professional actors stage the meeting where intellectual humility is hard, with the deadline pressure intact, before the real version arrives unannounced.
The honest test for any team reading the HBR piece this month is simple. Pick the most senior person in your last team meeting. Did they ask anyone, on purpose, what they were stuck on? Did they admit one thing they had been wrong about that week? If the answer to either is no, the learning-velocity gap is not a research finding. It is a forecast for your next quarter.
Book a free 30-minute consultation → or read about our research-backed approach.