How to evaluate experiential learning in L&D

← All articles

Experiential learning

How to evaluate experiential learning in L&D

Feedback forms show that people enjoyed it, not that anything changed. How I would evaluate experiential learning: clear signals, a patient check-in and honest stories.

Jonel

Jonel

Co-founder & CTO

·

·

6 min read

Sooner or later it happens in a budget meeting. Someone who was not in the room for the session, often someone from finance, asks a reasonable question: ‘How do we know it worked?’ And the person who organised it reaches for the only evidence they have, a folder of feedback forms full of top marks and warm comments. It is a good folder. It just answers a different question. It shows that people enjoyed the afternoon, not that anything changed afterwards.

Of all the questions that reach me at Uncommon Club, this is one I am always glad to get, because it means someone is taking the work seriously. Experiential learning is harder to measure than a compliance module with a quiz at the end, but it is far from unmeasurable. The trick is to measure the right things at the right time, without turning the session itself into an exam.

Evaluation starts before anything is booked

Evaluation that begins after the session is mostly guesswork. Before anything is booked, agree on two or three signals you would accept as evidence that something has changed, and make them observable. ‘Risks get raised in the weekly project meeting instead of in private afterwards’ is something a manager can notice. ‘Improved collaboration’ is not. If the session was designed around specific behaviours, as a good learning journey should be, most of this work is already done.

Then capture a starting point. It does not need to be elaborate. A short pulse question to participants, a few notes from managers on what they see today, or the relevant items from your last engagement survey will do. Without a baseline, any later improvement is an impression, and impressions do not survive budget meetings for long.

Kirkpatrick’s levels still work, with some adjustments

One of the most widely used frameworks in L&D is Donald Kirkpatrick’s four levels: reaction, learning, behaviour and results. It suits experiential learning well, as long as each level is bent a little to fit.

Reaction matters more here than people sometimes admit. If a group is not engaged, the experience never really happens, so enjoyment is not a vanity metric. But asking people to rate the day from one to five tells you very little. Better questions on the day are about memory and relevance: which moment do you think you will still remember in a month, how close was this to situations you face at work, and what is one thing you plan to try. One of the closing cards we use simply asks what surprised you about today, and the answers often point straight at the part of the session that carried the learning.

For learning, a knowledge quiz after an impro session would miss the point. A better approach is to ask people to rate their confidence on a few specific behaviours before and after, such as disagreeing with a senior colleague in a meeting, and to look at the quality of the plans they write at the end, which a good debrief makes room for. A specific, realistic plan is a decent sign that something has been understood.

A hand holds an Uncommon Club conversation card that asks: What surprised you about today?

The real evidence shows up a month later

Behaviour is where experiential learning should prove itself, and it is the level that gets skipped most easily, because by then the session is old news. Check in a month or two afterwards. Ask managers what they have noticed, using the signals you agreed at the start. Ask participants for examples rather than ratings: describe one situation in the last month where you used something from the session. Retrospectives and one-to-ones are natural places to collect this without adding yet another survey.

Results, the fourth level, need honesty. A single afternoon will rarely move a business metric on its own, and claiming that it did will cost you credibility the next time you ask for budget. What you can do is follow the indicators closest to the original problem, such as the engagement survey items on speaking up, and treat the session as one contribution among several rather than the sole cause. To my mind, one honest story is worth more than a number you cannot defend.

Stories carry what scores cannot

Numbers tell you whether something changed. Stories tell you how and why. Robert Brinkerhoff’s Success Case Method is a useful, lightweight way to gather them. You find a few participants who applied the learning most successfully and a few who applied it least, and you interview both groups. Brinkerhoff’s argument is that the difference often lies outside the training itself: in whether someone’s manager, team and workload left room to use it. That is the kind of insight no feedback form will ever give you.

A good interview prompt is simple: tell me about a moment since the session when something from it came back to you. Alongside the interviews, a few quieter signals are worth watching. Phrases from the session turn up unprompted in meetings and chats. People speak earlier in discussions, problems surface sooner, and colleagues from other teams start asking what happened in that room. None of these is proof on its own. Together, and in people’s own words, they make a far more convincing report than an average score.

Measuring the room changes the room

I would always advise against silent observers with clipboards during a session. People behave differently when they feel assessed, and the safety that makes experiential learning work disappears quickly. The behaviour you want to measure happens back at work anyway, so that is where to look for it.

The same caution applies to surveys. One short questionnaire on the day and one behavioural check-in later is plenty for most sessions. There is an old idea, usually credited to the economist Charles Goodhart, that once a measure becomes a target it stops being a good measure, and learning is not immune. If evaluation starts to push you towards sessions that are easy to score but shallow in effect, it is doing harm. The most valuable outcomes, a team whose members trust each other a little more or a leader who listens a little longer, are real even when they fit badly into a spreadsheet.

Honest evidence is enough

Good evaluation is mostly good planning: clear signals, a baseline and a patient check-in once the session has had time to settle. None of it needs special software or a research team.

When the budget meeting comes round again, it helps to arrive with one page rather than a folder. The two or three signals you agreed at the start. What managers have seen since. Two short stories, in participants’ own words, about a moment when the session came back to them. And one thing you would do differently next time. That page will not prove that an afternoon changed the whole company, and it should not try to. It will show that you took the question seriously, which in my experience is what the person asking it wanted to know.

Evaluation

Kirkpatrick Model

L&D

Jonel

About the author

Jonel

Co-founder & CTO

Jonel builds the systems behind Uncommon Club, the tools and technology that let a small team run sessions across cities without losing the personal touch that makes them work. He co-founded the club in Berlin in 2025, alongside Joost and Oktay.

Get in touch

Let’s talk through your programme.

Joost, co-founder of Uncommon Club
Oktay, co-founder of Uncommon Club

Tell us about your team, your occasion and what you want your people to walk away with. We’ll suggest the most impactful sessions and the artists who could lead them, tailored to your brief.

Want to send us an email? hello@uncommon.club

Get in touch

Let’s talk through your programme.

Joost, co-founder of Uncommon Club
Oktay, co-founder of Uncommon Club

Tell us about your team, your occasion and what you want your people to walk away with. We’ll suggest the most impactful sessions and the artists who could lead them, tailored to your brief.

Want to send us an email? hello@uncommon.club

Get in touch

Let’s talk through your programme.

Joost, co-founder of Uncommon Club
Oktay, co-founder of Uncommon Club

Tell us about your team, your occasion and what you want your people to walk away with. We’ll suggest the most impactful sessions and the artists who could lead them, tailored to your brief.

Want to send us an email? hello@uncommon.club