Kolb's learning cycle: how project managers learn from experience

Published 2026-07-02 · The research behind Project-simulator

Most project managers have heard that you learn from experience. But if that were entirely true, everyone with twenty years in the job would be a master of difficult conversations, stakeholder management and crisis leadership. They are not. The difference rarely lies in how much you have experienced, but in what you have done with those experiences.

The American educational researcher David Kolb put this into words back in 1984 in his book Experiential Learning. His model, often called Kolb's learning cycle, is still one of the most widely used theories of how adults learn from what they do.

What does the research say?

Kolb's core idea is that learning is not an event but a cycle with four stages. First comes the concrete experience: you do something, for example holding a tough conversation with a team member. Then comes reflection: you look back at what actually happened and how it felt. In the third stage you draw a conclusion, a small working rule or concept: perhaps that you tend to fill silences with apologies. In the fourth stage you actively test that conclusion in a new situation, which produces a new concrete experience, and the cycle begins again.

The point is that none of the stages is enough on its own. Experience without reflection is just repetition. Theory without testing becomes book knowledge that never settles into your hands. According to Kolb, learning arises precisely in the movement between doing, thinking it over, formulating and trying again.

Kolb built on earlier thinkers such as John Dewey and Kurt Lewin, and his model has since shaped everything from leadership programmes to medical education. That is also why serious forms of training almost always include a structured review afterwards, a so-called debrief, and not just the exercise itself.

What does this mean for you as a project manager?

First: schedule reflection, or it won't happen. After an important meeting or a difficult conversation, five minutes with three questions is enough: What happened? What did I do that worked or didn't work? What will I do differently next time? It sounds simple, but that is exactly the step that separates twenty years of learning from one year of experience repeated twenty times.

Second: turn your conclusions into small working rules. A vague feeling that the meeting went badly is impossible to practise. A concrete rule like "I will ask a question before giving my suggestion" can be tested as early as next week. Write the rules down, ideally in your project journal or after every retrospective.

Third: create opportunities to test. The cycle is only complete when you try your new conclusion in reality. Deliberately pick an upcoming meeting where you try the new behaviour, and evaluate afterwards. Then you have gone the whole way around, and that is when competence actually grows.

How to practise this

In everyday work, weeks can pass between opportunities to test a new conclusion, and the stakes are high when the moment finally comes. A simulator shortens the cycle to minutes: you carry out a conversation with an AI counterpart, get a review that points to your own wording, draw a conclusion and test it again immediately in the next round. That is Kolb's learning cycle in concentrated form. If you want to see how it feels, you can try Project-simulator free for a week.