Experience doesn't automatically produce learning. Every professor who has debriefed a simulation knows this: a team can play six rounds, make the same mistake in rounds two through six, and walk out of the room with the confident feeling that they now understand supply chains. They don't. What they have is a memory of doing something. Reflection is what turns that memory into understanding.
Six Rounds, One Debrief
The standard design for simulation-based learning goes something like this: participants play several rounds, results accumulate, and then there's a debrief at the end — usually the last thirty minutes of a session, usually when energy is lowest. In that debrief, the instructor tries to unlock what was learned across the whole exercise in the time that remains.
This design has a real problem. By the time the debrief happens, the details of round two are gone. Participants remember the dramatic moment in round four. They remember their final result. The texture of the decisions — why they chose what they chose, what information they were weighing, what they knew they were getting wrong but couldn't stop — has dissolved.
David Kolb's experiential learning model is well-known for a reason: it describes what we actually observe (Kolb, 1984). Concrete experience alone is not sufficient. It has to pass through reflection and conceptualisation before it becomes something a person can use. A simulation session that skips or rushes reflection is, at best, half the learning it could be — a point explored in depth in Practice to Learn: Importance, Factors and Reasons to Implement Simulations.
Closing the Loop, Round by Round
We built progress reflection reports into every Eureka Simulations exercise because the right moment to process a round is immediately after it ends — while the experience is still sharp. Three structured questions appear for each participant at the close of every round.
The questions are deliberately simple. What happened and what drove it? What does that tell you about the system you're operating in? What will you do differently next round? These are not essay prompts. A focused participant completes them in three to five minutes. The constraint is the point — it forces a choice about what actually mattered in that round, rather than allowing a stream of impressions to go unorganised.
This is different from the reflection journals and end-of-module essays that exist in many programmes. Those are disconnected from the simulation itself — written hours or days later, against a fading memory. What we've built is embedded. The reflection is part of the exercise, not an annotation of it.
What the Instructor Gets
When forty participants each complete a three-question reflection after each round, the instructor has access to something genuinely useful before the debrief: a diagnostic view of where the cohort's thinking actually is.
The AI progress summary reads all responses for a given round and surfaces patterns. Which teams are still explaining their results through external factors? Which ones are starting to identify the structural dynamics? Where is there disagreement within a team that hasn't yet surfaced in their decisions? This takes under a minute to generate, and it changes how an instructor enters the debrief room — with specific things to address rather than a general hope that the right topics come up.
It also means instructors don't need to read forty responses manually. The synthesis is done. The instructor can go directly to the responses that were flagged as outliers, or to the team whose summary suggests they've understood something no one else has yet. That's a different quality of preparation.
From Learning Experience to Learning Evidence
There's a practical dimension here that matters increasingly for institutions seeking accreditation through AACSB or EQUIS. Both bodies ask for documented evidence of learning outcomes — not just descriptions of activities, but demonstrations that specific competencies were developed in specific ways. A per-round reflection, tied to defined learning objectives, creates exactly that paper trail. Each submission is a timestamped, structured record of a participant's thinking at a specific point in a learning sequence.
We're not suggesting you run a simulation primarily to generate accreditation documentation. But the documentation that used to require a separate assignment — one more thing for participants to resent and instructors to read — now emerges naturally from the exercise itself. That's not an administrative benefit. It's a design improvement that happens to solve an administrative problem at the same time.
Want to see how reflection works in your next cohort?
Request a demo →
