Game-based learning in the workplace

Most workplace game-based learning is gamification wearing game-based learning's clothes. A compliance module with a progress bar and a badge at the end is still a compliance module. There is a quick test: take the scoring away. If nothing of the learning survives, the game was decoration.

We build these for a living and have watched them fail, so this page is written from the failure side. Mechanics are not neutral: each one rewards a particular behaviour, and if that is not the behaviour you want, the game will teach the wrong thing very efficiently.

Game-based learning, gamification and simulation are three different things

• Game-based learning. The game is the lesson. The situation, the decisions and the consequences carry the content. Remove the game and there is no session left.
• Gamification. Game mechanics wrapped around content that is not a game: points for finishing modules, streaks for logging in, a leaderboard on the intranet. It moves completion numbers. It rarely moves behaviour.
• Simulation. A faithful model of a real system, played straight. Good for procedure and for consequences that are expensive in real life. Usually humourless, which is fine, because a simulation is not trying to be fun.

They get sold interchangeably. The first question to ask a vendor is which one you are actually being offered.

What each mechanic actually rewards

Before you pick a mechanic, be honest about what it pays people to do. Participants are rational: they will optimise for whatever scores.

Leaderboards deserve a second look, being both the default and the most damaging. A public individual ranking tells the bottom half something they already suspected, in front of colleagues. If you want the energy without the damage, rank teams, reset every round and never publish a bottom. Where psychological safety is thin, skip it.

If you can win the game without doing the thing you are trying to teach, the game is teaching something else.

When game-based learning makes training worse

This is the section the platform vendors leave out. Six situations where a game is the wrong instrument:

1. There is one correct answer and a legal consequence. Policy, safe operating procedure, regulatory detail. Write a clear document, test it plainly, stop. Dressing it up as a quest wastes an afternoon.
2. The score is tied to performance review. The moment a game counts, people play the scoring system rather than the situation. You will get beautifully optimised behaviour and learn nothing about how anyone thinks.
3. The subject is raw. A recent restructure, a bereavement, a harassment case, a site accident. Gamifying something people are still hurt by reads as contempt, and the feedback form will say so.
4. Nobody will facilitate the debrief. Without someone to run the conversation afterwards, you have bought an expensive quiz. It is the commonest way a good game becomes a wasted day.
5. The mechanic contradicts the message. A collaboration workshop with an individual leaderboard. A psychological safety session where wrong answers lose points. People believe the mechanic, not the slide.
6. The group is small and the facilitator is good. Under about a dozen people, a skilled facilitator with a whiteboard and a hard question usually gets further than any software will.

There is a seventh, less comfortable one: sometimes you want a good day out rather than a training intervention. A fine thing to buy. Less fine when it is reported upward as development.

Four formats that hold up with adults

Each of these is runnable. Setup, numbers, time, what it surfaces, and the debrief question that makes it worth the room.

1. The decision game

• Setup. One situation that evolves in rounds. Teams get incomplete information and must commit to a decision before the next piece arrives. A facilitator releases each round so nobody races ahead.
• Group. Four to six per team, any number of teams.
• Time. Sixty to ninety minutes, plus thirty for the debrief.
• What it surfaces. Who escalates and who covers. Who verifies a claim and who takes it. Who goes quiet once the room sounds confident.
• Debrief question. At what point did you stop looking for information, and what made you stop?

2. The escape room or puzzle room

• Setup. A sequence of locked puzzles under a hard time limit. Include at least one puzzle that cannot be brute-forced by the loudest person, or you will simply watch that person work.
• Group. Six to eight. Above eight, half the room becomes an audience.
• Time. Forty-five to sixty minutes, then thirty for the debrief.
• What it surfaces. How a team splits parallel work, and how long a quiet person's correct answer takes to be heard.
• Debrief question. Who had the answer before the group accepted it, and what delayed it?

Be honest about this one: an escape room shows you the team rather than teaching it a skill. Good enough reason to run one, provided it is the reason you booked it.

3. The quiz used as a diagnostic

• Setup. Ten genuinely hard questions on the topic, answered in teams, each with a confidence rating. Score it so a confident wrong answer costs more than an admitted guess. Run it before you teach anything.
• Group. Three to five per team, any number of teams.
• Time. Fifteen minutes to play, thirty to argue about.
• What it surfaces. Confident wrongness, which is the only part of the topic worth your teaching time.
• Debrief question. Which answer were you most confident about and wrong on, and where did that confidence come from?

4. The hybrid hunt

• Setup. Teams solve something on a phone to earn access to a physical challenge, or run a physical challenge to earn the code that unlocks the next screen. Mental work on the device, physical work in the room, facilitators driving both.
• Group. Twenty to two hundred, in teams of eight to twelve, grouped into clusters.
• Time. Half a day.
• What it surfaces. Planning versus sprinting, and whether clusters share what they have found or hoard it.
• Debrief question. What did you decide not to do, and who made that call?

We use the hybrid format for large inductions. One of ours runs two hundred new joiners as sixteen teams across five levels: ciphers and deduction inside the app, physical ops off-screen, and a passcode from each puzzle that unlocks the next level. All four formats can be run with paper and a good facilitator. The reason we build custom learning games instead is scale and simultaneity: live scoring for two hundred people, and a reveal landing on every phone at the same second.

The numbers you will be shown, and why to distrust them

Somebody will show you a pyramid claiming people remember ten per cent of what they read and ninety per cent of what they do. Those percentages have no traceable source. They are usually credited to Edgar Dale's Cone of Experience from 1946, which contained no numbers at all; the figures were attached later and copied ever since. Doing does beat reading for most practical skills. You do not need invented arithmetic to say so.

The same applies to engagement uplift claims. When a vendor quotes a percentage, ask what it was measured against, over how long, and who funded it. The honest answer is usually that completion rates rose for a quarter.

Measure something you would act on instead. Name two or three observable behaviours before the session, then ask managers about them six weeks later. Kirkpatrick's model is the standard frame, and its uncomfortable point is that the levels easiest to collect are the least worth having.

How to specify one on Monday

1. Write the behaviour, not the topic. Escalates a bad number within a day is a behaviour. Ownership is a poster.
2. Decide what losing looks like. If no decision inside your game can go badly, you have built a slideshow with buttons.
3. Choose the mechanic last. Work back from the behaviour to the mechanic that pays for it. Most briefs run the other way and start with a leaderboard.
4. Write the debrief questions before the content. If you cannot write three questions the game would make interesting, it has nothing to reveal.
5. Name the facilitator and ring-fence the last thirty minutes. The debrief is the part that gets cut when the session overruns, which is like cutting the last chapter to save paper.
6. Pilot it with your most cynical team. They will find the exploit, tell you the story is thin, and save you from doing it in front of two hundred people.

Underneath it is an old idea: people learn from experience they have reflected on, not from experience alone. Kolb's cycle is the usual map of that, inside the broader family of experiential learning.

The debrief is the deliverable

The learning lands in the conversation afterwards, not on the scoreboard. An activity without a debrief is entertainment, and adults spot the difference immediately. The game exists to produce evidence: this team ignored the one person who was right, that team never asked where the number came from. A good debrief turns that into an argument about Monday.

If you want to feel the shape of it before commissioning anything, The Weekly Challenge is ten minutes, free, and ends in exactly the kind of debrief this page is arguing for.

Tour De Force runs this as live, experiential training for teams worldwide — online and in person. Talk to us, or play Gamified learning appsThe Weekly Challenge to see the method in ten minutes.