You have probably sat through a "scenario" that was really a quiz in costume: a cartoon character asks a question, you pick B, and a green tick appears. It does not feel like a real decision because nothing real happens. A good branching scenario feels different. You make a call, the situation changes, and you live with the result for a few screens. How do you build a branching scenario with decisions, consequences and feedback that feel real?

To build a branching scenario, start with one realistic decision your learners get wrong on the job, then map a short series of choices where each option leads to a believable consequence. Keep the map under control by letting branches rejoin at key moments instead of splitting forever. Show consequences before you explain them, so learners feel the result first and then get the reasoning. Decide early whether you are scoring the scenario or using it purely as practice, and test every path before anyone else sees it.

What Makes a Branching Scenario Different From a Quiz?

A branching scenario asks learners to apply judgment in a situation, while a quiz usually checks whether they remember something. In a quiz, each question stands alone and feedback is "correct" or "incorrect." In a scenario, each choice changes what happens next, so the learner sees a chain of cause and effect.

How Do You Build a Branching Scenario? – Designing Decisions, Consequences and Feedback That Feel Real - 3D Wireframe Animation

The open resource on branching scenarios from the University of Saskatchewan's Gwenna Moss Centre describes this well: learners make decisions at key points and see the consequences, which builds application rather than recall. That is the whole reason to go to the extra effort. If your goal is recall, a well-written quiz is cheaper and works fine.

There is also an emotional difference. A wrong answer on a quiz stings for a second. A wrong decision in a scenario, followed by an annoyed customer or a delayed shipment, sticks with people. That is exactly what you want, as long as it happens in a safe place.

How Do You Find a Decision Worth Practicing?

You find it by asking where people actually go wrong, not what the policy says. Sit down with your subject-matter expert and ask questions like "what do new people get wrong in their first few weeks?" and "what do experienced people do differently in the moment?" The answers point straight at decisions worth practicing.

A decision is worth a scenario when it has three features:

  • More than one option looks reasonable to someone who does not know better.
  • The wrong choice has a consequence the learner can understand and care about.
  • The right choice depends on noticing something in the situation, not just recalling a rule.

For the rest of this article, we will use a made-up example. A new help-desk agent gets a call from someone who says they are a manager locked out of their account before a big meeting. The caller is rushed, friendly and cannot answer the usual verification question. Should the agent reset the password? Plenty of new agents would, because helping feels like the job. That tension makes it a great scenario.

How Do You Map Branches Without Them Multiplying?

You stop branches from multiplying by letting paths rejoin at bottlenecks instead of giving every choice its own private future. The math gets out of hand fast otherwise. Three decisions with three choices each, branching fully, means 1 + 3 + 9 = 13 decision screens and 3 x 3 x 3 = 27 different endings. Nobody wants to write, review or test 27 endings.

A more manageable shape for our help-desk example looks like this:

  1. Decision 1: the caller asks for a reset. Options: reset it right away, follow the verification steps, or transfer the call to a supervisor.
  2. Each option shows its own consequence screen, then all three rejoin at Decision 2, where the caller pushes back and gets impatient.
  3. Decision 2 has three options and three consequences, then rejoins at Decision 3, a follow-up a day later.
  4. Decision 3 has three options, and the scenario ends on one of three endings based on the learner's overall choices.

Count it up: 3 decision screens, 9 consequence screens and 3 endings, so 15 screens in total instead of 13 decision screens plus 27 endings. The learner still feels that their choices matter, because each consequence is specific and the ending reflects the whole path. You get most of the realism at a fraction of the work.

Rejoining does not mean pretending the earlier choice never happened. A line of dialogue can carry it forward. If the learner transferred the call at Decision 1, the caller at Decision 2 might open with "I already waited for your supervisor once." That single sentence makes the shared screen feel personal, and it costs you a small variable and one alternate line instead of a whole new branch.

Keep the number of options steady as well. Three choices per decision is a comfortable default. Two can feel like a coin toss, and five usually means at least two of them are filler that nobody would pick in real life.

Sketch the map on paper or in a simple diagram before you write any dialogue. Give every screen an ID, such as D1, D1-A or E2, and draw the arrows. If you cannot draw it on one page, it is too big for a first version.

What Should Each Consequence Show the Learner?

Each consequence should show what happened because of the choice, inside the story of the scenario, before it says anything about right or wrong. In our example, if the agent resets the password without verification, the consequence screen might show a message the next morning: the real manager never called, and someone has accessed the account. Let the learner feel that for a moment.

Then add the coaching. Explain what the agent could have noticed (the caller dodged the verification question and leaned on urgency) and what the procedure is for. Short is better. Two or three sentences of reasoning land harder than a paragraph of policy.

Make the middle options interesting too. Transferring to a supervisor is not wrong, but the consequence might show a supervisor who is also busy and an agent who has learned nothing about handling the next call. Choices that are "okay but not best" teach nuance, and they stop the scenario from becoming an obvious one-right-answer puzzle.

Watch out for cartoon villains and obvious tells. If the "bad" option is written in a rude tone, everyone avoids it for the wrong reason. Write every option the way a real, well-meaning person might say it.

Should a Branching Scenario Be Scored?

A branching scenario should be scored only when you need a record of performance, and practice scenarios often work better without a score. When there is no score, learners feel free to try the risky option just to see what happens, and that exploration is valuable. You can still require them to reach the best ending before they continue.

If you do need a score, keep it simple and transparent. In our example, each decision could give 2 points for the best choice, 1 for a reasonable one and 0 for a risky one, for a maximum of 6. An ending for 5 to 6 points could show a confident agent, 3 to 4 a shaky one, and 0 to 2 a security incident. Tell the learner how scoring works before they start, so the result feels fair rather than mysterious.

Scenarios also pair well with more traditional checks. Use the scenario for judgment and a short set of questions for the facts behind it. If you are planning those, the article on question banks and assessments covers how to size and write them.

How Do You Test Every Path?

You test every path by working from your map, not from memory. Turn the map into a checklist with one row for each screen and each arrow. Then walk through the course and tick off every route. With 15 screens and a rejoining structure, a careful tester can cover every choice at every decision in a few passes.

Check three kinds of things as you go:

  • Logic: does each choice lead to the right consequence, and does the score or ending match the path taken?
  • Content: does each consequence still make sense no matter which earlier choice led here? Rejoined paths are where continuity errors hide.
  • Access: can a learner reach every choice with a keyboard, and do consequence screens have the same text alternatives as the rest of the course?

Then hand it to someone who has never seen it, ideally a person like your learners. Watch where they hesitate. If they pick an option for a reason you never imagined, that is useful information about how real people read your choices.

If you build in AuthorLearn Studio, the ready-made scenario or decision tree interaction gives you a starting structure, and once you insert it, it becomes ordinary objects, states, layers and triggers you can edit. That means you can reshape it to match your own map, including rejoining paths, rather than fitting your scenario to a fixed template.

A Branching Scenario Works When the Wrong Choice Teaches More Than the Right One

A strong branching scenario puts learners in a realistic moment, lets them make a believable mistake, and shows them what that mistake costs before explaining what to do instead. The right path confirms what they know. The wrong paths are where most of the learning happens, so give them your best writing.

Start with one decision people really get wrong, map it on a single page with rejoining paths, write consequences before coaching, and test every arrow. Your first scenario does not need ten decisions. Three good ones, each with a consequence that feels true, will teach more than a long quiz ever could.