How to gamify training without making it worse

We build custom learning games for clients, which mostly means we have watched this go wrong at close range: a scoring layer bolted onto a course, completion rates rising, nobody behaving differently on Monday. The method below is what we do instead. Behaviour first, mechanic last, debrief written before the game exists.

Step one: finish the sentence before you choose anything

Before a mechanic is discussed, finish this sentence: after this training, people will do more of ______, and we will notice because ______. Both blanks, out loud, in front of whoever is paying. If the first blank turns out to be finish the module, you do not need a game. You need a deadline and a manager who checks.

Then name the failure mode you are up against. Each takes a different instrument:

• They do not know. A knowledge gap. A clear document, a job aid or a quiz used as a diagnostic beats a game on cost and speed. Gamify this lightly, if at all.
• They know but do not practise. A repetition gap, and the one place points earn their keep. Reps, drills, spaced recall, streaks: the mechanic and the behaviour are the same shape.
• They know and cannot judge. A judgment gap, which no amount of decoration touches. It needs a scenario with real ambiguity, decisions made by teams, and someone to run the conversation afterwards.
• They know, they can judge, and the system punishes them for it. Not a training problem. Gamify this and you will have carefully taught a behaviour their manager penalises on Tuesday, which is worse than teaching nothing.

Most gamification disappointment is a mismatch between those four and the mechanic chosen. Get it right and the rest is craft.

The chooser: what to use, and what will quietly ruin it

The pattern, plainly: volume mechanics suit knowledge, speed mechanics suit emergencies, and judgment needs a mechanic that rewards reasoning, which nearly always means a human watching.

The rest of the sequence

1. Find the decision, not the content. Hunt through your material for the moment where a competent person could reasonably choose either way. That moment is your game. If no such moment exists, you have a reference document dressed as a course.
2. Choose the smallest mechanic that fits. One, from the table above. A second rarely doubles the effect; it splits attention, and people optimise for whichever number moves easiest. Points plus a leaderboard plus a timer is three incentives fighting over forty minutes.
3. Decide what advances the session: a facilitator or a clock. This decides what you are building. A clock makes it self-serve and removes any chance of intervening when a team goes down a productive wrong turn. A facilitator makes it a session rather than a product.
4. Write the debrief before the game. Three questions: what did you decide, what did you not know when you decided it, and where else do you decide that way. If the third one will not write, the design has no transfer in it yet.
5. Playtest with the two people most likely to break it. Your most competitive participant will find the shortcut; your most cynical one will tell you which part feels like school. Skip this and both get proved right in front of an audience.
6. Decide what losing costs. If the mechanic is competitive, half the room loses. Plan the last five minutes so they leave with something other than the experience of losing.

Write the debrief questions first. If you cannot write a good one, the design has no learning in it yet, and no amount of scoring will put any there.

Gamify a course you already run: the prediction round

You do not need software for this. Take a module you already deliver and steal thirty minutes from the front. Teams of four or five, eight to forty people, one slide and a shared sheet.

1. Predict before you teach. Put up five statements from the content, three true and two false, no hints. Teams commit in writing where everyone can see it. No hedging.
2. Score confidence, not correctness. Each team spreads ten confidence points across their five answers. Confidently wrong costs heavily; cautiously right earns little. Within two minutes teams are arguing about how sure they actually are, which is the point.
3. Teach the module unchanged. Same slides, same length. The difference is that everyone now has a stake in one part of it.
4. Reveal and settle up. Expect the heaviest losses on the statements the room found most obvious.
5. Debrief question: which one were you most confident about and most wrong about, and where else in your work do you make that call with exactly that much confidence?

It surfaces not ignorance, which is easy to fix, but confident wrongness, which is not. The teaching lands harder on people who have committed to a position and been caught. Cost: half an hour and a slide.

An activity without a debrief is entertainment. Run the prediction round without that last question and you have hosted a pub quiz. The questions in the order that works are in how to debrief an activity.

When not to gamify at all

The section platform vendors leave out. Six situations where scoring makes training worse:

1. When the subject is grave. A badge for finishing a module on harassment, bereavement or a fatal safety incident tells people the organisation regards the topic as a box. They will treat it accordingly, and be right to.
2. When the score is visible to a manager. As soon as a result reads as appraisal input, people stop playing and start performing. A game wired to a performance record is a test in a costume, and the room works that out before you do.
3. When you need people to admit something. Anything depending on psychological safety gets worse under scoring. You cannot incentivise candour with points, only make it safe and wait.
4. When the fix is a process, not a skill. If people skip the CRM because it has fourteen mandatory fields, a leaderboard for CRM hygiene produces fourteen fields of fast nonsense. Fix the form.
5. When nobody will facilitate. Without someone to run the debrief you have bought a quiz with better graphics. It is the most common way a good design dies.
6. When the group is small or senior. Below about a dozen people, a facilitator with a whiteboard gets further than any mechanic. And a room of directors opts out of anything resembling school faster than any audience you will meet.

The signals that tell you inside ten minutes

You do not need a survey to know a mechanic is failing:

• One person per team holds the pen throughout. The mechanic rewards speed, so the group has delegated to whoever is quickest.
• Teams ask whether something counts more than they discuss the problem. The score has become more legible than the objective.
• The room goes quiet after the first leaderboard. The ranking settled the race too early. Reset it, or drop it.
• The facilitator is enforcing rules rather than asking questions. There are too many rules, and they are eating the time the debrief needed.

Judging whether it worked, without inventing numbers

This topic is full of fabricated statistics, most famously the learning pyramid and its suspiciously round claim that we remember ten per cent of what we read and ninety per cent of what we do. Those percentages have no traceable source. Experiential learning stands up without them, and a supplier quoting a precise engagement uplift has a slide rather than a study.

Measure your own instead. Completion rate is an activity measure and should never be reported as an outcome. Pick one behaviour, agree what it looks like in the work, and check at ninety days. The Kirkpatrick model gives you honest levels; gamified sessions score beautifully at level one and nothing at level three unless the design went after it.

When decoration is not enough

If the behaviour you want is choosing well under pressure with incomplete information, no reward layer bolted onto existing content will reach it. The game has to become the training: a real situation, decisions made by teams, a facilitator controlling the pace, and a debrief that turns what happened into what changes.

That is what we build. HEARTWARE runs five rounds in ninety minutes on judgment in the AI era, for new joiners deciding what to trust, verify and own; teams question the characters in their own words rather than picking from a menu. Operation Keystone is a spy-thriller induction for two hundred new joiners in sixteen teams, deliberately hybrid: ciphers inside the app, physical challenges run off-screen. In both, every reveal lands on all the phones at the same second, so no team races ahead. More is on our gamified learning apps page, including when we tell people not to commission one.

If you would rather feel the difference than read about it, The Weekly Challenge is ten minutes of ours, free: a crisis week, seven decisions, and the debrief we build into paid work.

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.