Topic 4.1

Prototypes Take a Week. So You Skip Them.

AI builds structure in an hour. Enough to test before you commit.

⏱️ 12 minutes 📋 Prompt Templates ✓ Quality Checklist

The Prototype Problem

You have an idea for a branching scenario. Interactive decision points. Three possible outcomes.

To prototype properly: Map the flow. Write sample dialogue. Create placeholder screens. Test the logic. Get feedback.

Time required: 1 week minimum.

So you skip it. Go straight to full build. Discover halfway through that the branching doesn't work. Start over.

💡 The shift

AI generates prototype structure in 30-60 minutes. Enough to test the concept, get feedback, decide if it's worth building.

Why Prototypes Get Skipped

🚧 Four barriers to prototyping ▼
Barrier Reality
Time A week of work for something you'll throw away
Effort Feels like building twice—prototype, then real
Tools Prototyping tools have their own learning curve
Pressure "Just build it" beats "Let's prototype first"

Result: You build things that don't work. Waste weeks instead of days.

What AI Can Prototype

✓ AI Prototypes

  • Course structure and flow
  • Branching scenarios with decision points
  • Module outlines with estimated timing
  • Interactive activity descriptions
  • Assessment frameworks
  • Navigation and menu structures

✗ What AI Doesn't Do

  • Visual design
  • Actual interactivity
  • Final formatting

But you don't need those for a prototype. You need structure and logic.

The Basic Prompt

📋 Prototype generation prompt
Create a prototype for [course/module type]. Topic: [what it teaches] Audience: [who's learning] Goal: [what they'll be able to do] Format: [eLearning, microlearning, scenario, etc.] Include: - Overall structure - Estimated timing - Key decision points (if branching) - Assessment approach

Example:

Create a prototype for a branching scenario microlearning module. Topic: Handling difficult customer conversations Audience: Retail employees (new hires) Goal: Practice choosing appropriate de-escalation responses Format: Branching scenario, 5-7 minutes Include: - Scenario setup - 3 decision points - Multiple response options per decision - Consequences of each choice - Estimated timing per branch

What You Get

🎯 Example AI-generated prototype structure ▼

Module: Customer De-escalation Scenario
Estimated time: 6-7 minutes | Format: Branching scenario

Opening (0:30)

  • Context: Customer received wrong item, return deadline passed
  • Your role: Handle the situation
  • Stakes: Keep customer, follow policy

Decision Point 1 (1:00)
Customer: "This is unacceptable. I want a full refund NOW."

Options:

  • A) "I understand you're frustrated. Let me see what I can do."
  • B) "Our policy is clear. Returns are within 30 days."
  • C) "I'm sorry, but there's nothing I can do."

Branches:

  • Option A → Positive path, continues to Decision Point 2
  • Option B → Neutral path, harder to recover
  • Option C → Negative path, escalates further

[Structure continues through remaining decisions]

Outcomes:

  • Best: Customer satisfied with store credit solution
  • Moderate: Customer leaves unhappy but not escalated
  • Poor: Customer demands manager, writes negative review

Feedback points:

  • After each decision: Brief explanation of why response helped or hurt
  • End summary: What worked, what didn't, key principles

What AI Does Well vs. What It Gets Wrong

✓ AI Strengths

  • Generate logical flow structures
  • Create decision trees with consequences
  • Suggest realistic response options
  • Estimate timing for each section
  • Identify feedback opportunities

✗ AI Limitations

  • Overly complex branching (too many paths)
  • Unrealistic dialogue (too formal or generic)
  • Missing edge cases in logic
  • Unclear consequences for choices
  • Timing estimates sometimes off

💡 The Division of Labor

AI creates structure. You simplify, add realism, fix logic gaps, refine dialogue.

Prototype Types

📋 Four prototype types with time estimates ▼
Type AI generates You validate Time
Structure (course outline) Modules, topics, timing, flow Sequence, depth, completeness 15-20 min
Scenario (branching logic) Decision points, options, consequences Realism, complexity, learning value 30-45 min
Assessment (quiz framework) Question types, topics, difficulty progression Alignment, fairness, variety 20-30 min
Activity (interactive elements) Activity descriptions, instructions, feedback Feasibility, engagement, value 15-25 min

The Workflow

🔄 Six-step prototyping workflow (45-60 minutes total) ▼
  1. Write prototype prompt with context (5 min)
  2. AI generates initial structure (2 min)
  3. Review and identify gaps (10 min)
  4. Refine with follow-up prompts (15 min)
  5. Test logic — walk through paths (10 min)
  6. Get feedback from stakeholders (varies)

Total: 45-60 minutes vs. 1 week for traditional prototyping.

The Time Math

❌ Traditional prototyping

  • Map structure manually (4 hrs)
  • Write sample content (8 hrs)
  • Create mockups (12 hrs)
  • Test and revise (8 hrs)
  • Total: 1 week (32+ hours)

✅ AI-assisted prototyping

  • Prompt AI for structure (5 min)
  • Review and refine (25 min)
  • Test logic (10 min)
  • Gather feedback (15 min)
  • Total: 1 hour

Key Takeaways

  1. Prototype to test, not to build. AI creates structure in an hour—enough to validate before investing weeks.
  2. Catch problems early. Flawed branching logic discovered in prototype costs 1 hour. Discovered in build costs 1 week.
  3. Simplify AI's output. It will overcomplicate. Your job: cut paths, clarify logic, add realism.
  4. Use text, not visuals. Text-based prototypes are faster to create and easier to revise.

Try It Now

🎯 Your task:

Pick a project idea you've been hesitant to start. Generate a prototype with AI. Test the logic. Get feedback from one person.

The test: Did the prototype reveal a problem you would have discovered only halfway through building?

📥 Download: Prototyping prompts and validation checklists (PDF)

Ready-to-use templates for course structures, scenarios, and assessments.

Download PDF