Module 4: Idea Validation Techniques

Introduction

“The hardest part about startups is that you’ll have 100 reasons to believe your idea is great and ignore the 1 reason it’s not. That 1 reason is what kills you.” - YC Partner

This module teaches you how to validate ideas quickly and cheaply before investing significant time and resources.

The Validation Mindset

Core Principles

  1. Fall in love with the problem, not the solution
  2. Your opinion doesn’t count—only customer opinions matter
  3. Validation is about learning, not selling
  4. Speed matters more than perfection

The 18 Mistakes That Kill Startups (Paul Graham)

Most relevant to validation:

The Mom Test Framework

The Problem: People lie to be nice

Bad Question: “Would you use an app that helps you manage tasks better?” Good Question: “Tell me about the last time you struggled to manage your tasks. What did you do?”

Rules for Good Validation Questions

  1. Ask about specifics in the past, not hypotheticals
  2. Talk about their life, not your idea
  3. Listen more than you talk
  4. Follow up on emotions and frustrations

Example Validation Conversation

You: "How do you currently handle [problem]?"
Them: "Oh, I use spreadsheets mostly."
You: "Walk me through the last time you used them."
Them: "Well, last Tuesday I spent 2 hours trying to..."
You: "That sounds frustrating. How often does that happen?"

Validation Techniques

1. Customer Discovery Interviews

Goal: Understand if your problem hypothesis is correct

Process:

  1. Identify 20-30 potential customers
  2. Schedule 15-minute conversations
  3. Use the Mom Test framework
  4. Document patterns and insights

Template: Customer Interview Script

2. Landing Page Test

Goal: Test value proposition and demand

Components:

Success Metrics:

3. Concierge MVP

Goal: Deliver value manually before automating

Example: Uber started with calling black car services manually

Steps:

  1. Find 5-10 early customers
  2. Solve their problem manually
  3. Learn what really matters
  4. Only then build technology

4. AI-Assisted Validation

Goal: Use AI tools to accelerate and scale validation

Modern Tools:

When to use:

5. Pre-Sales Validation

Goal: Test if people will actually pay

Approaches:

Red Flag: If you can’t get 10 people to pay, you can’t get 1000

The Validation Canvas

Problem Validation

Solution Validation

Metrics That Matter

Common Validation Mistakes

1. Confirmation Bias

Symptom: Only hearing what you want to hear Fix: Have someone else do interviews

2. Building Too Soon

Symptom: “Let me just build a quick prototype” Fix: Can you test without building anything?

3. Asking Leading Questions

Bad: “Don’t you hate how hard it is to…?” Good: “Tell me about your experience with…”

4. Ignoring Bad News

Symptom: “They just don’t get it yet” Fix: Bad news is good news—it saves you time

5. Too Broad Target Market

Symptom: “Everyone could use this” Fix: Start with a specific, desperate segment

YC’s Validation Wisdom

From Paul Graham’s “Startups in 13 Sentences”

  1. Pick good cofounders
  2. Launch fast - validation starts with real users
  3. Let your idea evolve - validation shapes your product
  4. Understand your users - validation never stops
  5. Better to make a few users love you - validate depth over breadth

From Startup School

Practical Exercises

Exercise 1: Problem Safari

  1. Spend 2 hours in places where your target customers are
  2. Observe without talking—what problems do you see?
  3. Document 10 specific observations
  4. Which problems appear most painful?

Exercise 2: Five Whys Analysis

  1. Start with a problem statement
  2. Ask “why” five times to get to root cause
  3. Validate each level with customer interviews
  4. The deepest “why” is often the real opportunity

Exercise 3: Build Your Validation Plan (6-Week Cycle)

Create a modern validation sprint:

Validation Metrics & Benchmarks

B2C Validation (2025 Standards)

B2B Validation (2025 Standards)

Resources & Tools

Tools for Validation (2025)

Key Takeaways

  1. Validation saves you years of wasted effort
  2. Customer opinions > Your opinions
  3. Test the riskiest assumptions first
  4. Bad news early is a gift
  5. Validation is ongoing, not one-time

Next Steps

✓ Complete customer interview template ✓ Schedule 10 interviews this week ✓ Create simple landing page ✓ Define your kill criteria ✓ Share learnings with the community

“The only way to win is to learn faster than anyone else.” - Eric Ries

← Previous: Understanding Your Users Next: MVP Development →