calendar
Edited on October 28, 2026

Experiential Learning Activities: 12 Workplace Activities and How to Debrief Them

12 experiential learning activities for teams at work, each matched to a stage of Kolb's cycle with the debrief questions to ask afterwards.

Alex Hey
Alex Hey
Chief Operating Officer
Key takeaways

Experiential learning activities are structured tasks where people do something real or realistic, then reflect on what happened, draw a rule from it and try that rule on the next piece of work.

  • A meta-analysis of 46 samples found an average debrief benefit, described by its authors as about 25% in effectiveness. That is not a promised gain for a particular team.
  • Choose an activity by the stage your team skips, such as reflection or trying the rule on real work.
  • Debrief what went well as well as what failed, and book a follow-up on real work before the session ends.
HR

Train teams faster and more effectively with AI

A manager runs a role play at the Friday team meeting. Two people act out an awkward customer call, everyone laughs, and the session ends on time. On Monday the same customer calls go the same way, because nobody stopped to ask what happened in the role play or what to do differently.

I'm the COO at Coursebox and I manage most of our Perth team. In a team our size there is no HR layer between me and the conversation, so a training session that produces nothing I can point at a week later is a session I won't run twice. When I hire, I don't ask people how they'd handle a customer; a colleague and I play the customer and watch. What I check a week later is whether anything changed, and that depends on the debrief more than on the activity. So pick the activity by the stage of Kolb's learning cycle your team keeps skipping (the experience, the reflection, the rule or the try-out), and write the debrief questions before you run it. Coursebox builds AI course tools for employee training, and when I built this kind of module with it, its first debrief kept none of the answers; the fix is in the last section.

Experiential learning activities are structured tasks where people do something real or realistic, then reflect on what happened, draw a rule from it and try that rule on the next piece of work. Below are 12 for teams at work, each with a run sheet and three debrief questions.

What are experiential learning activities?

Experiential learning activities are structured tasks where people do something real or realistic, then reflect on what happened, draw a rule from it and try that rule on the next piece of work.

The Association for Experiential Education describes the approach as engaging learners in "direct experience and focused reflection". The reflection half is the part that gets dropped. The National Society for Experiential Education put it plainly in its eight principles of good practice for experiential learning activities: "Reflection is the element that transforms simple experience to a learning experience."

Many workplace designs borrow David Kolb's cycle: a concrete experience, reflective observation, abstract conceptualisation (turning the story into a rule), and active experimentation (trying the rule). Our guide to the four stages of Kolb's cycle covers the stages, the learning styles debate and the limits. This post is about the activities: which one to run, how to run it, and what to ask afterwards.

Kolb’s cycle connects concrete experience, reflective observation, abstract conceptualisation and active experimentation, with a role play, review, job aid and supported work task as examples.

Benefits of experiential activities, and what the debrief adds

The benefit you want from an experiential activity is transfer: people doing something differently on the job. The research on workplace training points to practice that resembles the job, errors made safely during training, and a structured look back afterwards.

A debrief is that look back: the structured conversation after the activity about what was supposed to happen, what did happen, and what to keep or change. Without one, people keep whatever lesson they happened to take from the experience, which may be the wrong one. In Tannenbaum and Cerasoli's meta-analysis of team and individual debriefs, debriefs "systematize reflection, discussion, and goal setting to promote experiential learning", and the effect was similar in simulated and real settings and in medical and non-medical groups.

About 25%
average improvement in effectiveness when people were debriefed, compared with control groups, across 46 samples and 2,136 people (d = .67)
Source: Tannenbaum and Cerasoli (2013), Human Factors, meta-analysis

That figure is an average across studies. The authors also found that results were better when the debrief's focus lined up with what was being measured. In practice that means debriefing the skill you trained, and leaving how the session felt for another time.

For the sessions in this guide, plan a structured debrief. The second comes from a field study in an army navigation course: debrief what went well as well as what went wrong.

A person in a khaki shirt checks a compass on their wrist while holding a folded paper map outdoors
cottonbro studio / Pexels

Case study: Navigation training reviews in the Israel Defense Forces

Two companies of soldiers (98 men aged 18 to 20) in an elite unit were in their navigation course: repeated exercises scored on points reached, pace and map use, each followed by a one-to-one review of about 15 minutes with an experienced commander. The unit's reviews usually focused on failures.

  • What they did: In the second week, one company's commanders switched to reviews of failures and successes. Soldiers named the positive and negative parts of the exercise, discussed the decisions behind both, and ended by noting three things to improve or keep unchanged. The other company kept failure-focused reviews of the same length.
  • Evidence and result: In Ellis and Davidi's 2005 study (76 performance scores analysed), both companies improved during the week, and the company reviewing successes and failures improved significantly faster, even as the exercises grew harder. In a subsample of 20 soldiers interviewed, mental models of successes started out thinner than those of failures, and the gap closed over the course.
  • Limitation: Whole companies were assigned, there was no group without reviews, the company reviewing successes started from a lower score, and the comparison covered three exercises in one week. Navigation has clear right and wrong answers, which a sales call or a handover often does not.
  • Lesson: Keep one question about what went well in every debrief, and ask what decision produced it, so the good move gets repeated on purpose.

How to choose an activity: start from the stage your team skips

Pick the activity from the symptom you see at work. Each stage of the cycle fails in a recognisable way:

StageSymptom you see at workActivities below
Concrete experienceThey know the policy and freeze on the call1 to 3
Reflective observationThe same mistake every month, and nobody talks about it4 to 6
Abstract conceptualisationEveryone does it differently and nobody can say why7 to 9
Active experimentationA great session, then no change on Monday10 to 12

Use the stages as a planning lens. Activities can involve several stages, and people may enter through an existing experience, a question or a model. Check what support is missing rather than diagnosing a stage from one symptom or always starting at the same point.

12 experiential learning activities for the workplace

Each run sheet uses the same labels, so you can lift one straight into a session plan.

1. Micro role play with a hard stop

  • Stage: Concrete experience.
  • Time: 15 minutes for three rounds.
  • Group: Threes: one learner, one person playing the other side, one observer.
  • Set-up: Write one opening line from the hardest moment of a real conversation, such as "So you're charging me more for the same thing?"
  • Run it: The role-player delivers the line, the learner responds, and the observer stops the round at 90 seconds, mid-exchange. Rotate roles.
  • Debrief questions: "What did you say in the first ten seconds?" / "Where did the conversation turn?" / "What would you say instead at that point?"
  • Watch for: Rounds that run to a neat ending; the stop is what makes people look at one moment.

Our role play examples for workplace training give full briefs for both sides, and the customer service role play scenarios give 25 more opening situations.

2. Error-allowed simulation

  • Stage: Concrete experience.
  • Time: 30 to 45 minutes.
  • Group: Pairs at one screen or one workstation.
  • Set-up: A safe copy of the real system or process: a training login, a test order, a sample file. Tell people their job is to break it.
  • Run it: Give the pair a realistic task and a few mistakes to start with, such as an expired card or a missing field, then let them try their own. They note what the system does each time.
  • Debrief questions: "Which mistake surprised you?" / "How would a customer or colleague notice it?" / "How would you recover?"
  • Watch for: Trainers who step in to stop the errors; the errors are the content.

This activity borrows from error management training, where learners explore and are told to make errors. In Keith and Frese's meta-analysis of error management training, both active exploration and explicit encouragement to make errors were effective elements.

d = 0.56
effect of error management training on transfer after training, across 24 studies with 2,183 people; the overall effect was d = 0.44
Source: Keith and Frese (2008), Journal of Applied Psychology, meta-analysis

The larger effect showed up on transfer after training, which is the situation you are training for. For conversations rather than systems, a branching version works the same way; see scenario-based learning or branching scenario software for building one.

3. Shadow, then swap: the workplace version of an internship

  • Stage: Concrete experience.
  • Time: Two consecutive instances of a real task.
  • Group: One learner, one experienced colleague.
  • Set-up: Pick a task that repeats within a day, such as a quote, a handover or a close-out.
  • Run it: The learner watches the colleague do the first one and writes down what they notice. Then they swap: the learner does the next one while the colleague watches and stays quiet unless something is about to go wrong.
  • Debrief questions: "What did you do differently from what you watched?" / "Where did you hesitate?" / "What did your colleague do that you didn't notice the first time?"
  • Watch for: The colleague taking over during the swap.

It fits cross-training employees and the steps in how to train someone at work.

4. After-action review on a real recent event

  • Stage: Reflective observation.
  • Time: 20 minutes, within a week of the event.
  • Group: Everyone who was involved.
  • Set-up: Choose one recent event that went well or badly, and write four questions on the board: what was supposed to happen, what happened, why was there a gap, and what do we keep or change?
  • Run it: Answer the four questions in order. The facilitator writes; the most senior person speaks last.
  • Debrief questions: "What went well that we should keep?" / "Who owns the one change?" / "When do we check it worked?"
  • Watch for: Jumping to the fourth question before the second is agreed.

The army navigation study above tested a close cousin of this format: naming what went well and badly, and the decisions behind each. For a fuller reflective model with feelings and evaluation, see Gibbs' reflective cycle.

5. Fishbowl with an observer checklist

  • Stage: Reflective observation.
  • Time: 25 minutes.
  • Group: Two people do the task in the middle; everyone else observes.
  • Set-up: A checklist of three named behaviours, such as "asks what the customer needs before quoting".
  • Run it: The pair runs the task for five minutes. Observers tick each behaviour when they see it and write down the exact words used.
  • Debrief questions: "Which behaviour did you see, and what were the words?" / "Which did you not see?" / "What would you have said?"
  • Watch for: Observers grading the people instead of noting the behaviours.

Our guide to observation checklists covers how to write the behaviours.

6. Timeline rebuild

  • Stage: Reflective observation.
  • Time: 30 minutes.
  • Group: The team that ran the process.
  • Set-up: A whiteboard with three columns: time, what happened, why.
  • Run it: The team rebuilds a process that went wrong, minute by minute, filling the first two columns before anyone writes in the third.
  • Debrief questions: "At what point could someone have seen it coming?" / "What information was missing then?" / "What check would have caught it?"
  • Watch for: Names in the "why" column; write down the condition that allowed the mistake.

7. Two-case comparison

  • Stage: Abstract conceptualisation.
  • Time: 20 minutes.
  • Group: Small groups of three or four.
  • Set-up: Two short examples of the same task, one that went well and one that did not, such as two call recordings or two written quotes.
  • Run it: Groups compare the cases, identify differences and propose a rule to test. Separate observations from explanations; two cases alone cannot establish which difference caused the result.
  • Debrief questions: "What is the rule?" / "When would the rule not apply?" / "What would you need to see to know someone is following it?"
  • Watch for: Rules written as values ("be customer focused") instead of actions.

8. Write the job aid

  • Stage: Abstract conceptualisation.
  • Time: 30 minutes.
  • Group: The team that will use it.
  • Set-up: Their own notes from an earlier debrief.
  • Run it: The team drafts the one-page checklist or script they will use on the job, in their own words, and tests it against one real example.
  • Debrief questions: "What would you cut?" / "Which line would you forget under pressure?" / "Where will this live so you see it at the moment you need it?"
  • Watch for: Losing the learners' reasoning during revision. Keep their contribution, but have the responsible expert approve any safety, legal or operating instructions before the job aid is used.

9. Teach-back

  • Stage: Abstract conceptualisation.
  • Time: 10 minutes.
  • Group: Pairs: one person who was at the session, one who missed it.
  • Set-up: The rule from activity 7 or 8.
  • Run it: The person who was there explains the rule in under two minutes. The listener asks one "what if" question about a case it does not cover.
  • Debrief questions: "Which part was hardest to explain?" / "What did the 'what if' expose?" / "Does the rule need another line?"
  • Watch for: Explanations that recite the slide and skip the reason.

10. Trainee-written scenarios

  • Stage: Active experimentation.
  • Time: 40 minutes.
  • Group: The team, including their manager.
  • Set-up: Each person writes the hardest version of the situation they expect to face this month, in three sentences.
  • Run it: Swap scenarios and practise them in pairs, using the rule or script from activities 7 to 9.
  • Debrief questions: "Which scenario was hardest, and why?" / "Did the rule hold?" / "When will you meet this situation for real?"
  • Watch for: The manager sitting out; they practise too.

Both choices follow Taylor, Russ-Eft and Chan's review of 117 behaviour modelling training studies: transfer to the job was greatest when, among other things, practice included trainee-generated scenarios and the trainees' superiors were also trained.

11. Stretch assignment with an after-event review

  • Stage: Active experimentation.
  • Time: One real task, plus a 15-minute review.
  • Group: One person and their manager.
  • Set-up: A real task slightly above the person's level, such as leading a client meeting. Book the review before the task starts.
  • Run it: The person does the task. In the review, they answer the four questions from activity 4 about their own performance.
  • Debrief questions: "What did you decide in the moment, and why?" / "What went well that you would repeat?" / "What will you try differently next time?"
  • Watch for: Reviews that slip off the calendar when the task goes fine.

DeRue and colleagues followed 173 people across four developmental experiences over nine months and found that structured after-event reviews had a positive effect on leadership development. Our post on coaching vs training vs mentoring covers who should run that review.

12. One change, rerun

  • Stage: Active experimentation.
  • Time: One run of a recurring process, plus a check two weeks later.
  • Group: The team that runs the process.
  • Set-up: A recurring process, such as a handover, a stand-up or a quote, and one change agreed in a debrief.
  • Run it: Rerun the process with exactly one change and compare it with the last run. Two weeks later, ask each person: "Where did you use it, and what happened?"
  • Debrief questions: "What did the change do?" / "What did it cost?" / "Do we keep it?"
  • Watch for: Changing three things at once, so nobody knows which one worked.

A worked example: one skill through all four stages

This example is hypothetical. A newly hired account manager has to call a long-standing customer whose annual plan price is going up at renewal. They know the reasons for the increase. On the phone, they over-apologise and offer a discount nobody approved. The goal is that on their next two renewal calls they explain the change before they offer anything. Each activity feeds the next: the role play produces the moment, the review finds the line where it went wrong, the job aid turns that into a rule, and the real calls test it.

StageActivityThe debrief question that did the workWhat they leave with
Concrete experience1. Micro role play: a colleague opens with "So you're charging me more for the same thing?" and stops the round at 90 seconds"What did you say in the first ten seconds?"Their own words, written down
Reflective observation4. After-action review of the role play"At what exact line did you start negotiating against yourself?"The line where the concession began
Abstract conceptualisation8. Write the job aid: three lines to say before any concession"Which line would you forget under pressure?""Here's what changed in the plan." "Here's what you keep." "Here are your options."
Active experimentation11. The next two real renewal calls, with a 15-minute review after each"What went well that you would keep?"Two real calls to compare and a script revised through two sampled calls

This is the hiring move I described at the start, with the step most sessions skip added on: the review and the real calls afterwards.

Experiential learning activities for remote and hybrid teams

Six of the 12 work on a video call with small changes: the micro role play (1), the after-action review (4), the timeline rebuild (6), the two-case comparison (7), the teach-back (9) and trainee-written scenarios (10). Use breakout rooms for the threes and pairs. The fishbowl (5) and shadow, then swap (3) work best in a room unless the task itself is done on a shared screen. The rest (2, 8, 11, 12) happen on real systems or real work, so they run the same way wherever people sit.

On video, the facilitator calls the hard stop out loud ("Stop there"), because people miss a silent timer. Observers type the exact words into the chat as they hear them.

When experiential activities go wrong

Experiential activities fail in four common ways:

  • Novices dropped into an unguided task. The title of Kirschner, Sweller and Clark's 2006 paper names experiential teaching among the approaches it argues against: "Why Minimal Guidance During Instruction Does Not Work". Give new staff the rule or a model first, then the experience. Our guide to discovery learning covers when exploration helps.
  • The activity teaches the wrong lesson. Without a debrief, the person who froze in a role play may conclude "I'm bad at this" rather than "I need an opening line".
  • The debrief turns into blame. If the review hunts for a culprit, people hide the next mistake. Keep names out of the "why" and ask what went well first.
  • Games with no skill target. A quiz night or an outdoor day can build goodwill, but if nobody can name the skill it practises, it is a social event. Training games work when each one maps to a skill.

Building experiential activities into a course with Coursebox

Our team keeps a sample of course requests typed into Coursebox, and I searched it for activities and debriefs.

About 1 in 5
course requests typed into Coursebox asked for activities, exercises, scenarios or hands-on practice; none of the 104 used the word "debrief"
Source: Coursebox course requests (PostHog), 104 English requests, 90 days to 7 October 2026, keyword match then read

That is one 90-day sample of English requests, and the activity count is a keyword-based estimate (about 23 of 104). It says nothing about what people built afterwards. It did make me want to see what the course agent does when someone asks for the debrief explicitly.

So I asked it for a 30-minute practice module for new account managers, using the renewal call from the worked example, in four sections: an AI role play where the customer opens with "So you're charging me more for the same thing?", a debrief with three open-text questions about that role play, a one-page three-line script, and an assignment to use the script on the next two real calls and answer the same three questions after each. No multiple choice anywhere.

It built four sections and four pages, 30 minutes, in my order, with no multiple choice. The debrief questions pointed back at the role play ("Quote or closely recall the line, then explain what made it self-negotiation"), and the final assignment repeated all three questions under "Call 1" and "Call 2". The role play itself was not on the page: section 1 was two text blocks telling the learner to open the course AI tutor, which the agent said was set up to play the customer with my opening line.

The debrief was a custom HTML widget. Its page said entries were not saved and should be copied elsewhere to keep them. Reflection can still support learning without a stored answer, but this design needed a record the learner could revisit and a manager could review. The widget did not supply that record.

I sent one message to the AI editor: make section 2 an open-text assignment with exactly these three questions, and delete the HTML widget and any quiz on that page. It replied that it had. I left section 1 on the course AI tutor to keep the test to one fix. The edit also added a three-row marking rubric I had not asked for, so delete it if you want the first debrief ungraded. I did not take the reply as proof, so I reloaded the page: "Review your first response" now holds an assignment block with "At what line did you start negotiating against yourself?", "What did the customer actually object to?" and "What went well that you would keep?", the widget is gone, and the final assignment still asks the same three questions after each call.

I verified the blocks and questions after reload, not the tutor conversation, learner submission, stored response or marking workflow. Section 1 still depends on opening the separate course tutor, and its claimed role-play configuration was not tested in a conversation. Those learner checks remain before team use.

Start with one activity

Experiential learning activities work when the experience is followed by a debrief and the debrief is followed by real work. Pick the stage your team skips from the table above, choose one of the three activities for it, write the three debrief questions before the session, and book the follow-up on real work before people leave the room. If you only change one thing, add "What went well that you would keep?" to every debrief you already run.

The debrief cards below put all 12 run sheets on printable cards, with a stage picker, a "what went well" line, a commitment line and a two-week follow-up box.

Experiential activity debrief cards

One card for each of the 12 activities in this article, with the stage, time, group, set-up, how to run it, the three debrief questions, a "what went well" line, a "next time I will" line and a two-week follow-up box, plus a one-page stage picker. It is a planning aid and gives no score. It is an editable Word file, and no email is required.

Frequently Asked Questions

Experiential learning activities are structured tasks where people do something real or realistic, then reflect on what happened, draw a rule from it and try that rule on the next piece of work. Role plays and after-action reviews are common at work.

Five common ones: a 90-second role play of the hardest line in a real conversation; an error-allowed simulation in a safe copy of a system; an after-action review of a recent event; writing a job aid from the team's own debrief notes; and a stretch assignment with a 15-minute review booked before it starts.

Ask four questions in order: what was supposed to happen, what happened, why there was a gap, and what you will keep or change. Include what went well as well as what failed, keep names out of the "why", and end with one change and a date to check it on real work.

The 70-20-10 rule says people learn about 70% from experience on the job, 20% from other people and 10% from formal training. It is a practitioner rule of thumb rather than a research finding, so use it as a reminder to plan on-the-job practice. Our guide to the 70-20-10 learning model covers where it came from.

In Kolb's experiential learning theory they are concrete experience, reflective observation, abstract conceptualisation and active experimentation.

Alex Hey

Alex Hey

Chief Operating Officer

Alex Hey is a technical marketer with 10 years of hands-on growth and digital marketing experience. He cofounded Coursebox together with Travis Clapp who he met online when looking for a cofounder to build a globally impactful software product. He works in the Perth, Western Australia office as the COO. He built the world's first automated funnel for patient recruitment before building Coursebox. After launching Coursebox in mid 2023, he drove +400k organic sign ups for Coursebox. Alex leads the marketing and digital growth of Coursebox and works with the product development team bringing a passion for data-driven product decision and exceptional user experience. Alex is determined to build the world's easiest way to build the perfect course.