What is scenario-based learning?

Presenting a realistic situation and asking what to do, rather than testing recall of a rule. It targets recognition, which is what actually fails in practice — people usually know the rule but do not recognise the moment it applies.

What makes a good training scenario?

Difficulty in the situation rather than the wording. Someone who understands the material should get it right; someone who has only memorised the rule should plausibly get it wrong.

Where do I find realistic scenarios?

From whoever handles incidents, complaints or escalations in your organisation. Ask what people get wrong — you will usually get three or four recurring situations, already the right difficulty.

Should e-learning scenarios use branching?

Only where the lesson is that decisions compound and an early minor choice determines how bad things get later. Branching multiplies writing and review cost, and a well-written single-decision scenario usually does the job.

How should I write scenario feedback?

As if speaking to someone who just made that choice. Explain why a wrong answer was tempting and what distinguished the correct one — and explain the correct answer too, since a lucky guess teaches nothing.

Course Design

Scenario-Based Learning: Teaching Judgement, Not Just Rules

October 21, 2026

Ask anyone who has investigated a workplace incident and they will tell you the same thing: the person involved usually knew the rule. They could have recited it. What they did not do was recognise, at the moment it mattered, that this was the situation the rule was about.

What problem do scenarios actually solve?

The gap between knowing a rule and applying it. Conventional training teaches the rule and tests recall of the rule, which confirms only that the words went in. A scenario presents a situation and asks what to do, which tests the thing that actually fails in practice — recognition.

This is why scenario-based questions frequently produce much worse scores than recall questions on the same content, and why that is a useful finding rather than a problem with the questions.

What makes a scenario worth including?

Difficulty in the right place. A good scenario is hard because the situation is ambiguous, not because the question is tricky or the wording is convoluted.

The test: someone who genuinely understands the material should get it right, and someone who has only memorised the rule should plausibly get it wrong. If both get it right, it is a recall question in costume. If both get it wrong, it is unfair.

Where do good scenarios come from?

Not from imagination. The best source is things that have actually happened, which is why the most valuable conversation in any course project is with the person who handles the incidents, complaints or escalations.

Ask them what people get wrong. You will typically get three or four recurring situations, and those are your scenarios — already realistic, already the right difficulty, and already the things worth training. Anonymise them properly, change enough detail that nobody is identifiable, and check with whoever owns the material before publishing.

What does a well-built scenario look like?

Short setup, genuine dilemma, plausible options:

Keep the setup brief. Two or three sentences. Long scenarios become reading comprehension tests, and the learner's failure is then about attention rather than judgement.
Make every option defensible. If one answer is obviously the responsible one, the learner picks it without thinking. Every distractor should be something a reasonable person might do.
Include the tempting shortcut. The wrong answer that is wrong because it is convenient is the most instructive option you can offer.
Explain both directions. Say why the right answer is right and why each wrong answer is tempting. That second part is where most of the learning happens.

Should scenarios branch?

Sometimes, and less often than the enthusiasm for branching suggests. Branching is genuinely valuable when the point is that decisions compound — the early choice that seemed minor determines how bad the later situation becomes. That is a real lesson and it cannot be taught in a single question.

It is expensive, though. Every branch multiplies the writing and the review, and a badly-branched scenario where three paths converge on the same outcome teaches the opposite of the intended lesson. For most content, a well-written single-decision scenario with good feedback does the job at a fraction of the cost.

How many should a course have?

Fewer than you might expect, placed where judgement is genuinely required. A course with two excellent scenarios and clear content is better than one with a scenario after every module, most of them thin.

The natural placement is at the end of a topic, once the learner has the material, and at the end of the course as a synthesis where several topics interact. That last one is often the most useful thing in the whole course, because real situations do not arrive labelled by module.

How do I write the feedback?

As if you were talking to someone who just made that choice, not as if you were marking it. "That would be reasonable if the request had come through the normal channel — the thing that makes this one different is that it arrived directly" teaches. "Incorrect. See policy section 4." does not.

Feedback on the correct answer matters too, and is usually neglected. Someone who guessed correctly learns nothing unless you tell them why it was right.

The uncomfortable part

Good scenarios will make your completion statistics look worse and your training considerably better. A course where everyone scores full marks first time has told you nothing about anyone. If nobody ever gets a question wrong, the questions are not doing any work — and neither, probably, is the course.