Validation Interview

Validation Interview

Overview

Validation interviews are not the same as discovery interviews. In discovery, you explore the problem. In validation, you present a proposed solution and gauge whether it works. This skill covers concept testing, prototype testing, and usability interview techniques.

When to Use

• You have a prototype or concept to test with real users
• You need to validate assumptions before committing to build
• You want to catch usability issues before shipping
• You're practicing continuous discovery — testing solutions as you design them
• Don't use for: early-stage discovery (use customer-interviews skill), quantitative validation (use A/B tests), or sales demos (different goal)

Instructions

1. Prepare the Interview

Define the goal: What specific assumptions are you testing?

Choose the method:

• Concept test: Show a description or mockup. Test desirability and comprehension.
• Prototype test: Let them use an interactive prototype. Test usability and flow.
• Wizard of Oz: Pretend the product works. Test the core value hypothesis.

Write the script:

• Opening: "I'm going to show you something we're thinking about building. I want your honest reaction — negative feedback is more valuable than positive."
• For each concept: "Here's the idea. What do you think this does?"
• For prototypes: "I'd like you to try completing [task]. Please think aloud as you go."
• Closing: "If this existed today, would you use it? Why or why not?"

2. Conduct the Interview

Do:

• Let users explore freely before explaining
• Ask "What would you expect to happen next?"
• Pay attention to body language and hesitation
• Encourage criticism: "What's wrong with this?"

Don't:

• Lead the user through the interface
• Explain what they missed (until after testing)
• Get defensive about your design
• Ask "Do you like it?" (production bias)
• Over-explain the concept (if it needs explanation, that's a finding)

3. Watch for Warning Signs

• Polite praise: "That's cool!" → probe for specifics
• Feature requests: Building features doesn't validate the core problem
• Confusion: If they don't understand what it does, that's a signal
• Solution indifference: "I could see this being useful" = not a strong signal

4. Synthesize Results

For each validation interview:

1. Comprehension check: Did they understand the concept without explanation?
2. Desirability check: Would they use it? What would it replace?
3. Usability check: Could they complete tasks independently?
4. Caveats: What conditions would make them NOT use it?

5. Make a Decision

Patterns across 5+ interviews:

• ✅ 4/5 understood + expressed desire → proceed
• 🔄 Mixed signals → iterate prototype and re-test
• ❌ 3/5 confused or indifferent → reconsider the solution
• ❌ 5/5 didn't understand → start over with clearer framing

Sample Output

6 concept tests for a "team updates" feature:

• ✅ Comprehension: 5/6 understood the concept without explanation
• ✅ Desirability: 4/6 said they'd use it "today"
• ⚠️ Usability: 3/6 tried to find it in settings, not the project view
• 🔄 Action: Iterate placement before usability testing. Re-test with 3 more users.

Common Pitfalls

1. Confirmation seeking — If you're looking for people to validate your idea, you'll hear what you want
2. Leading the witness — "Don't you think this would save time?" = useless answer
3. Ignoring the non-verbal — Facial expressions, pauses, and sighs tell you more than words
4. One interview, one decision — A single validation interview is an anecdote. 5+ is a pattern
5. Falling in love — If you can't imagine the answer being "kill it," don't interview. You're not ready to hear the truth

Verification Checklist

• [ ] Interview goal tied to specific assumptions
• [ ] Script avoids leading questions and solution pitching
• [ ] Participants represent the target segment
• [ ] 5+ interviews completed per concept
• [ ] Results synthesized into go/iterate/kill recommendation
• [ ] Findings shared with the team before building

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: uthumany
• Source: uthumany/uthy-legacy-os
• License: MIT
• Homepage: https://uthumany.github.io/uthy-legacy-os/

Install and usage instructions live in the source repository linked above.