The Part of Action Learning Most Programs Leave Out

‘Action learning’ appears in 63 percent of U.S. executive leadership programs, and a survey cited by Marquardt et al. (2010) found 77 percent of learning executives named it their top driver of leadership bench strength. What that same research points out, and what most program designers seem to move past quickly, is that most of what gets labeled ‘action learning’ is missing the mechanism that makes it work. Organizations have adopted the name and format while quietly removing the part that produces the development.

What the Format Actually Requires

‘Action learning’ is a development approach built on four specific conditions: a real organizational problem with genuine stakes, a small group of participants with diverse perspectives, real actions taken on that problem between sessions, and a trained coach whose sole job is to ask questions rather than give answers. Marquardt et al. (2010) drew the line deliberately: outdoor adventures, case study discussions, and problem solving groups without structured reflection don’t qualify, even when marketed as action learning.

I used to think the coaching component was adjacent to the real work, a kind of supportive structure around the learning rather than the engine of it. The research changed that view. The coach isn’t a feature; the coach is the mechanism that converts problem solving into leadership development. Without someone whose only role is to slow the group down and ask what participants are noticing, assuming, or avoiding, the group optimizes for the problem’s solution and the learning from how they solved it goes unexamined.

That’s a real distinction. Most organizational problems get resolved by whoever knows the most or speaks most confidently. ‘Action learning’ is specifically designed to develop the people who might not know the most, by forcing them to surface their reasoning and question each other’s assumptions under real conditions.

What the Evidence Shows

Volz-Peacock et al. (2016) reviewed action learning implementation across hundreds of organizations worldwide and found that when run with a trained coach, the approach develops leadership capabilities while simultaneously solving urgent organizational problems. The learning doesn’t compete with the work. The work is the curriculum.

Leonard et al. (2010) reviewed 21 empirical studies measuring the actual impact of action learning and identified consistent evidence for three outcomes: development of broad leadership skills, particularly collaborative leadership and coaching ability; improved capacity to find integrative solutions to conflict situations; and high sensitivity to specific conditions, with skilled coaching, a safe environment for honest exchange, and genuine group diversity appearing in nearly every highly successful implementation.

That last condition, group diversity, tends to get underweighted in most program designs. When every participant comes from the same function, you’ve limited the quality of the questioning before the first session starts. The challenges that surface assumptions most effectively tend to come from people who think through genuinely different lenses, and that can’t be designed into a group of five people doing essentially the same kind of work.

Where Programs Break Down

Three things happen when organizations adopt the action learning name but not the structure.

The first is substituting simulated problems for real ones. Simulated problems remove the stakes, and without stakes, participants optimize for appearing capable rather than for learning from the process. The honest conversation about what the group doesn’t know, what’s politically difficult, and what might fail only happens when the outcome actually matters to the people in the room.

The second is treating the small group as a project team. A project team shares ownership of an outcome, while an ‘action learning set’ shares ownership of each member’s learning process. The conversations that flow from those different purposes look nothing alike.

The third is running sessions without a coach trained in the specific method. A skilled facilitator manages process. A skilled action learning coach manages reflection. Asking what the group produced is a facilitation question. Asking what each person noticed about how they respond under pressure is a coaching question. The research is consistent: the second type is where the leadership development lives.

What to Change

Start with the problem, not the participants. The right design question is: which live organizational challenges would genuinely benefit from perspectives that span functions, while also developing the specific capabilities the participant group most needs to build? That overlap identifies the right problem. Starting from a competency model and working backward to a fabricated scenario produces the first failure mode above.

Invest in the coaching role before anything else. Volz-Peacock et al. (2016) are direct: action learning with a trained coach works. Without one, the approach tends toward sincere group discussion that doesn’t transfer. If the organization doesn’t have internal capacity for this yet, building it or partnering externally before the first cohort is worth doing before the program design conversation.

Build in action between sessions. Participants should be doing something in the organization related to the problem, not just preparing to discuss. The leadership learning that transfers comes from reflecting on actual decisions made, not from rehearsing what might be done.

Build in diversity deliberately. Span functions, levels, and if possible, geographies. The diversity isn’t just good practice here; it’s the source of the questioning quality that makes reflection possible in the first place.

---

One question worth raising with whoever designs your leadership programs: when did the last cohort last do something real in the organization, get it partly wrong, and have a structured conversation about what that revealed about how they actually think under pressure? If the program doesn’t include that, it may not be action learning in the sense the evidence supports.

If you’re thinking through what this kind of design could look like for your team, this is exactly the work we take on at Peak Potential. You can find us at peakpotentialconsulting.com/services/team-workshops or reach out directly.

Frequently Asked Questions

What is ‘action learning’ and how does it differ from traditional leadership programs?

‘Action learning’ is a structured development method built around four conditions: a real organizational problem with genuine stakes, a small diverse group, real actions taken between sessions, and a trained coach focused on reflection rather than instruction. Marquardt et al. (2010) defined the distinction deliberately: activities using the label without coached reflection, including simulations, case studies, and project groups, lack the component that produces transferable development.

Why is the coach role so central to action learning’s effectiveness?

Leonard et al. (2010) found the presence of a skilled coach to be one of the most consistent predictors of success across 21 empirical studies on action learning outcomes. The coach’s role isn’t to advise on the problem. It’s to ask questions that slow participants down long enough to notice what they’re assuming, avoiding, or reacting to under real conditions. That kind of structured reflection doesn’t emerge naturally in a problem solving group without someone creating the condition.

Does action learning produce leadership skills that transfer beyond the program?

The evidence reviewed by Volz-Peacock et al. (2016) suggests yes, when implemented correctly. The reason transfer tends to be stronger than in many program formats is that the learning happens within a real work context from the start. Skills developed while navigating genuine organizational complexity tend to remain accessible when similar complexity appears in settings the program didn’t specifically prepare for.

How should an organization choose which problems to use for action learning?

Look for live challenges that genuinely need perspectives from multiple parts of the organization, where the outcome matters enough to generate honest conversation, and where the thinking required maps to capabilities the participant group most needs to develop. Marquardt et al. (2010) noted that what distinguishes action learning from other experiential formats is precisely its insistence on real actions affecting real organizational outcomes, so selecting a problem the organization would work on anyway is almost always stronger than constructing one for the program.