πŸ“‘ Table of Contents

  1. What Is Lean Validation?
  2. The Build-Measure-Learn Loop
  3. Types of MVPs
  4. Customer Development Framework
  5. Hypothesis Testing
  6. Validation Experiments
  7. Metrics That Matter
  8. When to Pivot vs Persevere
  9. Lean Validation Tools
  10. Case Studies
  11. Common Mistakes
  12. Frequently Asked Questions

What Is Lean Validation?

Lean validation is a systematic approach to testing business ideas with minimal investment before committing significant resources.

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                    LEAN VS TRADITIONAL                          β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  TRADITIONAL APPROACH                                           β”‚
β”‚  ────────────────────────────────────────────────               β”‚
β”‚  Idea β†’ Plan β†’ Build (6-12mo) β†’ Launch β†’ Discover no one wants β”‚
β”‚                                                                 β”‚
β”‚  Cost: $50,000-500,000+                                        β”‚
β”‚  Risk: Very High                                                β”‚
β”‚  Time: 6-18 months                                              β”‚
β”‚                                                                 β”‚
β”‚  LEAN APPROACH                                                  β”‚
β”‚  ────────────────────────────────────────────────               β”‚
β”‚  Idea β†’ Test β†’ Learn β†’ Iterate β†’ Build what's validated        β”‚
β”‚                                                                 β”‚
β”‚  Cost: $0-5,000                                                 β”‚
β”‚  Risk: Minimal                                                  β”‚
β”‚  Time: 2-8 weeks per cycle                                      β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

🎯 Core Principles of Lean Validation

Principle Description Why It Matters
Validated Learning Test assumptions, not just build features Avoid building wrong things
Minimum Viable Product Smallest thing that tests hypothesis Conserve resources
Build-Measure-Learn Rapid iteration cycles Fast feedback loops
Pivot or Persevere Data-driven direction changes Avoid sunk cost fallacy
Innovation Accounting Actionable metrics over vanity metrics Real progress measurement

πŸ’‘ Why Lean Validation Works

The problem it solves:

  • ❌ 42% of startups fail due to "no market need"
  • ❌ Most products are built on untested assumptions
  • ❌ Traditional planning can't predict customer behavior

The lean solution:

  • βœ… Test assumptions before major investment
  • βœ… Get real customer feedback early
  • βœ… Iterate based on data, not opinions

The Build-Measure-Learn Loop

πŸ”„ The Core Engine of Lean Validation

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                BUILD-MEASURE-LEARN LOOP                         β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚                        IDEAS                                    β”‚
β”‚                          β”‚                                      β”‚
β”‚                          β–Ό                                      β”‚
β”‚                    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”                                  β”‚
β”‚                    β”‚  BUILD  β”‚ ← Minimum viable experiment      β”‚
β”‚                    β””β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”˜                                  β”‚
β”‚                         β”‚                                       β”‚
β”‚                         β–Ό                                       β”‚
β”‚                    PRODUCT/MVP                                  β”‚
β”‚                         β”‚                                       β”‚
β”‚                         β–Ό                                       β”‚
β”‚                    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”                                  β”‚
β”‚                    β”‚ MEASURE β”‚ ← Actionable metrics             β”‚
β”‚                    β””β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”˜                                  β”‚
β”‚                         β”‚                                       β”‚
β”‚                         β–Ό                                       β”‚
β”‚                       DATA                                      β”‚
β”‚                         β”‚                                       β”‚
β”‚                         β–Ό                                       β”‚
β”‚                    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”                                  β”‚
β”‚                    β”‚  LEARN  β”‚ ← Validated learning             β”‚
β”‚                    β””β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”˜                                  β”‚
β”‚                         β”‚                                       β”‚
β”‚                         β–Ό                                       β”‚
β”‚              β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”΄β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”                            β”‚
β”‚              β”‚                     β”‚                            β”‚
β”‚              β–Ό                     β–Ό                            β”‚
β”‚          PIVOT               PERSEVERE                          β”‚
β”‚     (Change course)      (Double down)                          β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ“Š Loop Components Explained

1. BUILD Phase

Goal: Create the minimum experiment needed to test your hypothesis.

Build Type Time Investment Best For
Landing page 1-3 days Testing demand
Mockup/prototype 1-5 days Testing usability
Concierge MVP 1-2 weeks Testing workflow
Functional MVP 2-6 weeks Testing core value

Key principles: - Build the smallest thing possible - Focus on learning, not perfection - Speed matters more than polish

2. MEASURE Phase

Goal: Collect actionable data about customer behavior.

Metric Type Examples Why It Matters
Acquisition Sign-ups, visits Are people finding you?
Activation First actions, engagement Do they understand value?
Retention Return visits, continued use Is it sticky?
Revenue Purchases, upgrades Will they pay?
Referral Shares, invites Will they recommend?

Key principles: - Measure actions, not opinions - Focus on conversion metrics - Track cohorts, not aggregates

3. LEARN Phase

Goal: Extract validated learnings from data.

Questions to answer: - [ ] Did the hypothesis prove true? - [ ] What surprised us? - [ ] What should we test next? - [ ] Should we pivot or persevere?


Types of MVPs

πŸ› οΈ MVP Spectrum: Low-Fidelity to High-Fidelity

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                      MVP FIDELITY SPECTRUM                      β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  LOW FIDELITY ◄───────────────────────────────────► HIGH       β”‚
β”‚                                                                 β”‚
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”   β”‚
β”‚  β”‚Smoke   β”‚  β”‚Explainerβ”‚  β”‚Conciergeβ”‚  β”‚Wizard  β”‚  β”‚Functionalβ”‚ β”‚
β”‚  β”‚Test    β”‚  β”‚Video   β”‚  β”‚MVP     β”‚  β”‚of Oz   β”‚  β”‚MVP      β”‚   β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”˜   β”‚
β”‚                                                                 β”‚
β”‚  Cost:      Low ──────────────────────────────────► High       β”‚
β”‚  Time:      Hours ────────────────────────────────► Weeks      β”‚
β”‚  Fidelity:  Fake ─────────────────────────────────► Real       β”‚
β”‚  Learning:  Interest ─────────────────────────────► Behavior   β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ“‹ MVP Types Detailed

1. Smoke Test / Landing Page MVP

What it is: A page describing your product before it exists.

Metric What It Tells You
Page visits Demand/interest level
Sign-up rate Perceived value
Email opens Continued interest
Pricing clicks Willingness to pay

How to build: 1. Create landing page (Carrd, Webflow, HTML) 2. Describe the problem and solution 3. Add clear call-to-action 4. Drive traffic (ads, communities) 5. Measure conversions

Best for: Testing demand before building anything.


2. Explainer Video MVP

What it is: A video demonstrating your concept.

Famous example: Dropbox launched with just a demo video that got 75,000 signups overnight.

How to build: 1. Script the value proposition 2. Create simple demo (screen recording or animation) 3. Post with signup form 4. Measure signups and shares

Best for: Products that need to be seen to be understood.


3. Concierge MVP

What it is: Manually delivering your service before automating it.

Aspect Manual Concierge Automated Product
Scale 1-10 customers Unlimited
Cost per customer High Low
Learning Deep Surface
Speed to launch Immediate Weeks/months

How to run: 1. Find 5-10 potential customers 2. Deliver service manually 3. Observe and interview 4. Identify automation opportunities

Best for: Services, workflows, anything with complex user needs.


4. Wizard of Oz MVP

What it is: An interface that looks automated but is human-powered behind the scenes.

Famous example: Zappos started by photographing shoes at stores, listing them online, then buying and shipping when orders came in.

How to run: 1. Build the front-end experience 2. Manually fulfill on the back-end 3. Customers don't know it's manual 4. Validate before building real tech

Best for: Testing if customers want the experience, not just the concept.


5. Functional MVP

What it is: A working product with minimum features.

The "minimum" part is crucial:

Include Exclude
Core value proposition Nice-to-haves
One way to accomplish goal Multiple paths
Basic UI that works Polished design
Essential metrics Advanced analytics

Best for: When lower-fidelity MVPs have validated demand.


🎯 Choosing the Right MVP Type

Validation Question Best MVP Type
"Is there demand?" Smoke test / Landing page
"Do they understand it?" Explainer video
"What do they really need?" Concierge
"Will the experience work?" Wizard of Oz
"Will they use and pay?" Functional MVP

Customer Development Framework

πŸ‘₯ The Four Steps to Epiphany

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚              CUSTOMER DEVELOPMENT PROCESS                       β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  STEP 1: CUSTOMER DISCOVERY                                     β”‚
β”‚  └── "Do I understand the problem?"                            β”‚
β”‚      β€’ Identify target customers                                β”‚
β”‚      β€’ Understand their problems                                β”‚
β”‚      β€’ Test problem assumptions                                 β”‚
β”‚                                                                 β”‚
β”‚  STEP 2: CUSTOMER VALIDATION                                    β”‚
β”‚  └── "Will they pay for my solution?"                          β”‚
β”‚      β€’ Test product assumptions                                 β”‚
β”‚      β€’ Get first paying customers                               β”‚
β”‚      β€’ Validate business model                                  β”‚
β”‚                                                                 β”‚
β”‚  STEP 3: CUSTOMER CREATION                                      β”‚
β”‚  └── "Can I reach more customers?"                             β”‚
β”‚      β€’ Find scalable channels                                   β”‚
β”‚      β€’ Build demand generation                                  β”‚
β”‚      β€’ Optimize conversion                                      β”‚
β”‚                                                                 β”‚
β”‚  STEP 4: COMPANY BUILDING                                       β”‚
β”‚  └── "Can I build an organization?"                            β”‚
β”‚      β€’ Scale operations                                         β”‚
β”‚      β€’ Build team                                               β”‚
β”‚      β€’ Formalize processes                                      β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

🎀 Customer Interview Techniques

The Mom Test Rules

Problem: People lie to be nice. Your mom will say your idea is great even when it's not.

Solution: Ask questions about their life, not your idea.

Bad Question ❌ Good Question βœ… Why
"Would you use this?" "How do you handle this today?" Past behavior > hypotheticals
"Do you like this feature?" "Walk me through the last time..." Specifics > generalities
"Would you pay $X?" "What have you tried? What did you pay?" Real actions > intentions
"What would make this better?" "What's the hardest part?" Problems > solutions

Interview Structure

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                  CUSTOMER INTERVIEW FLOW                        β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  1. CONTEXT (5 min)                                            β”‚
β”‚     β€’ Tell me about your role/situation                        β”‚
β”‚     β€’ What's a typical day/week look like?                     β”‚
β”‚                                                                 β”‚
β”‚  2. PROBLEM EXPLORATION (15 min)                               β”‚
β”‚     β€’ What's the hardest part about [topic]?                   β”‚
β”‚     β€’ Walk me through the last time that happened              β”‚
β”‚     β€’ Why is that a problem? What does it cost you?            β”‚
β”‚                                                                 β”‚
β”‚  3. CURRENT SOLUTIONS (10 min)                                 β”‚
β”‚     β€’ What have you tried to solve this?                       β”‚
β”‚     β€’ What do you use now? What do you like/dislike?           β”‚
β”‚     β€’ Have you paid for solutions? How much?                   β”‚
β”‚                                                                 β”‚
β”‚  4. REACTION (optional, 5 min)                                 β”‚
β”‚     β€’ Show concept (only after understanding problem)          β”‚
β”‚     β€’ Watch for genuine excitement vs politeness               β”‚
β”‚     β€’ Ask for specific commitment (signup, payment)            β”‚
β”‚                                                                 β”‚
β”‚  5. REFERRAL (2 min)                                           β”‚
β”‚     β€’ Who else should I talk to about this?                    β”‚
β”‚     β€’ Can you make an introduction?                            β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ“ Interview Tracking Template

Customer Problem Intensity Current Solution Willingness to Pay Key Insight
Person A High (mentioned 5x) Spreadsheets $50/mo Needs automation
Person B Medium Free tool X Skeptical Happy with status quo
Person C High Manual process $100/mo Time is critical

Pattern recognition: 2 of 3 have high problem intensity and willingness to pay. Focus on that segment.


Hypothesis Testing

πŸ”¬ From Assumptions to Experiments

Every startup is built on assumptions. Lean validation is about testing them.

πŸ“‹ Assumption Mapping

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                   ASSUMPTION CATEGORIES                         β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  1. PROBLEM ASSUMPTIONS                                         β”‚
β”‚     β€’ The problem exists                                        β”‚
β”‚     β€’ It's painful enough to solve                              β”‚
β”‚     β€’ People are aware they have it                             β”‚
β”‚     β€’ It's a priority to solve                                  β”‚
β”‚                                                                 β”‚
β”‚  2. CUSTOMER ASSUMPTIONS                                        β”‚
β”‚     β€’ We know who the customer is                               β”‚
β”‚     β€’ We can reach them                                         β”‚
β”‚     β€’ They have budget for solutions                            β”‚
β”‚     β€’ They have authority to buy                                β”‚
β”‚                                                                 β”‚
β”‚  3. SOLUTION ASSUMPTIONS                                        β”‚
β”‚     β€’ Our solution solves the problem                           β”‚
β”‚     β€’ It's better than alternatives                             β”‚
β”‚     β€’ Customers will understand it                              β”‚
β”‚     β€’ We can build it                                           β”‚
β”‚                                                                 β”‚
β”‚  4. BUSINESS MODEL ASSUMPTIONS                                  β”‚
β”‚     β€’ Customers will pay $X                                     β”‚
β”‚     β€’ We can acquire customers for <$Y                          β”‚
β”‚     β€’ They'll stay long enough (LTV > CAC)                      β”‚
β”‚     β€’ The market is big enough                                  β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

🎯 Hypothesis Template

Use this format to write testable hypotheses:

We believe [CUSTOMER SEGMENT]
will [EXPECTED BEHAVIOR]
because [REASON/VALUE PROPOSITION]

We will know we're right when [MEASURABLE OUTCOME]

Example:

We believe remote team managers
will sign up for a daily standup tool
because it saves 30+ minutes per day on coordination

We will know we're right when 20% of landing page visitors sign up

πŸ“Š Prioritizing Hypotheses

Factor Weight Score 1-5
Risk if wrong 40% How fatal to business?
Confidence 30% How sure are you it's true?
Testability 30% How easy to get clear signal?

Priority = (Risk Γ— 0.4) + ((5 - Confidence) Γ— 0.3) + (Testability Γ— 0.3)

Test highest priority (highest risk, lowest confidence, most testable) first.


Validation Experiments

πŸ§ͺ Experiment Design Framework

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                    EXPERIMENT DESIGN CARD                       β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  HYPOTHESIS: _______________________________________________   β”‚
β”‚                                                                 β”‚
β”‚  EXPERIMENT TYPE: β–‘ Landing Page  β–‘ Interview  β–‘ Prototype    β”‚
β”‚                   β–‘ Concierge     β–‘ Ad Test    β–‘ Pre-sale     β”‚
β”‚                                                                 β”‚
β”‚  SUCCESS METRIC: __________________  TARGET: ________________  β”‚
β”‚                                                                 β”‚
β”‚  TIMELINE: Start ________  End ________  Duration: ________   β”‚
β”‚                                                                 β”‚
β”‚  RESOURCES NEEDED: ________________________________________    β”‚
β”‚                                                                 β”‚
β”‚  DECISION CRITERIA:                                            β”‚
β”‚  β€’ If metric > target β†’ Proceed to ________________________   β”‚
β”‚  β€’ If metric < target β†’ Pivot to __________________________   β”‚
β”‚                                                                 β”‚
β”‚  RESULT: ________________________________________________      β”‚
β”‚                                                                 β”‚
β”‚  LEARNINGS: ____________________________________________       β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ“‹ Common Experiment Types

1. Landing Page Test

Component What to Test
Headline Problem resonance
Value prop Benefit clarity
CTA button Signup conversion
Pricing Willingness to pay
Social proof Trust signals

Success metrics: - Conversion rate > 5% (good) - Conversion rate > 10% (excellent) - Email list signups


2. Ad Test

Element What It Validates
Click-through rate Messaging resonance
Cost per click Competitive landscape
Landing conversion Product-market fit signal
Cost per signup Customer acquisition cost

Minimum test budget: $100-500 per variant


3. Pre-sale Test

The ultimate validation: Getting money before building.

Pre-sale Type How It Works
Crowdfunding Kickstarter, Indiegogo
Founding member Annual/lifetime at discount
Waitlist deposit Refundable deposit for priority
Service first Sell manual service, then automate

Why it works: - Credit cards don't lie - Paying customers are committed - Proves complete conversion funnel


4. Fake Door Test

What it is: Add a feature button that doesn't work yet, measure clicks.

How to run: 1. Add button/link for potential feature 2. When clicked, show "Coming soon" with signup 3. Measure click rate and signup rate 4. Build features with proven demand

Ethics: Be transparent. Don't frustrate users.


Metrics That Matter

πŸ“Š Vanity vs. Actionable Metrics

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                VANITY VS ACTIONABLE METRICS                     β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  VANITY METRICS (Avoid)           ACTIONABLE METRICS (Use)     β”‚
β”‚  ────────────────────────         ────────────────────────      β”‚
β”‚  Total signups                    Weekly active users           β”‚
β”‚  Page views                       Conversion rate               β”‚
β”‚  Total downloads                  Retention rate                β”‚
β”‚  Total users                      Revenue per user              β”‚
β”‚  "Registered users"               Paying customers              β”‚
β”‚  Social followers                 Referral rate                 β”‚
β”‚  App store rating                 NPS from actual users         β”‚
β”‚                                                                 β”‚
β”‚  WHY VANITY METRICS FAIL:                                       β”‚
β”‚  β€’ Can go up while business is failing                         β”‚
β”‚  β€’ Don't show if you're building the right thing               β”‚
β”‚  β€’ Can be gamed easily                                         β”‚
β”‚  β€’ Don't indicate willingness to pay                           β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

🎯 The One Metric That Matters (OMTM)

At any stage, focus on one primary metric:

Stage OMTM Why
Problem validation Interview insights Understand problem depth
Solution validation Signup conversion Measure interest
Product validation Activation rate Do they get value?
Business validation Paying customer % Will they pay?
Growth Referral rate Sustainable growth?

πŸ“ˆ Pirate Metrics (AARRR)

Metric Question Example KPIs
Acquisition How do users find you? Signups, visits, sources
Activation Do they have great first experience? Completed onboarding, first action
Retention Do they come back? Week 1 retention, DAU/MAU
Revenue How do you make money? Conversion to paid, ARPU
Referral Do they tell others? Invite rate, NPS, K-factor

πŸ“Š Cohort Analysis

Why cohorts matter: Aggregate data hides problems.

Week Cohort A (Jan) Cohort B (Feb) Cohort C (Mar)
Week 1 100% 100% 100%
Week 2 45% 52% 58%
Week 4 22% 28% 35%
Week 8 12% 18% 24%

Insight: Retention is improving with each cohort. Product changes are working.


When to Pivot vs Persevere

πŸ”„ The Pivot Decision Framework

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                   PIVOT OR PERSEVERE                            β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  PIVOT SIGNALS πŸ”΄                    PERSEVERE SIGNALS 🟒       β”‚
β”‚  ────────────────                    ──────────────────         β”‚
β”‚  β€’ Experiments consistently fail     β€’ Metrics improving        β”‚
β”‚  β€’ No one will pay                   β€’ Customers asking for moreβ”‚
β”‚  β€’ Problem isn't painful enough      β€’ Word-of-mouth happening  β”‚
β”‚  β€’ Can't reach customers             β€’ Clear path to growth     β”‚
β”‚  β€’ Losing to simpler alternatives    β€’ Retention is strong      β”‚
β”‚  β€’ Team losing conviction            β€’ Team energized           β”‚
β”‚                                                                 β”‚
β”‚  QUESTIONS TO ASK:                                              β”‚
β”‚  1. Are we making progress on metrics that matter?             β”‚
β”‚  2. Are customers genuinely excited (not just polite)?         β”‚
β”‚  3. Is there a clear path to 10x improvement?                  β”‚
β”‚  4. What would we do differently if we started over?           β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ“‹ Types of Pivots

Pivot Type What Changes Example
Zoom-in Feature becomes product Instagram started as check-in app, photo filters became the product
Zoom-out Product becomes feature Your product becomes part of bigger offering
Customer segment Same product, different audience Shift from enterprise to SMB
Customer need Same audience, different problem Discovered bigger pain point
Platform Application to platform Or platform to application
Business model How you charge Subscription to usage-based
Channel How you sell Direct to through partners
Technology How you build Different tech stack/approach

🎯 Healthy Pivot Process

  1. Document learnings - What did experiments teach us?
  2. Identify options - What pivots are possible?
  3. Validate new direction - Apply same rigor to pivot
  4. Commit fully - Half-pivots waste time
  5. Reset metrics - Start measuring fresh

Lean Validation Tools

πŸ› οΈ Tool Stack by Category

Landing Pages

Tool Price Best For
Carrd Free-$19/yr Simple pages fast
Webflow $12-36/mo Design flexibility
Unbounce $90/mo A/B testing
Typedream Free-$15/mo Notion-like building

Customer Research

Tool Price Best For
Calendly Free-$12/mo Scheduling interviews
Zoom Free-$15/mo Video interviews
Typeform Free-$25/mo Surveys
UserTesting $49/video User testing

Analytics

Tool Price Best For
Google Analytics 4 Free Traffic basics
Mixpanel Free-$25/mo Product analytics
Amplitude Free-$995/mo User behavior
PostHog Free-$450/mo Open source analytics

Prototyping

Tool Price Best For
Figma Free-$15/mo Design prototypes
Maze Free-$99/mo Prototype testing
Marvel Free-$16/mo Quick prototypes

Advertising

Platform Min Budget Best For
Google Ads $50+ Search intent
Facebook/Instagram $20+ Interest targeting
Reddit Ads $5/day Niche communities
LinkedIn Ads $100+ B2B targeting

Case Studies

πŸ“˜ Case Study 1: Dropbox

Problem: Cloud sync was hard to explain.

Traditional approach would have been: Build full product, then launch.

Lean approach: 1. Created 3-minute demo video 2. Posted to Hacker News 3. 75,000 signups overnight 4. THEN built the product

Lesson: Validate demand before building.


πŸ“˜ Case Study 2: Buffer

Problem: Unclear if people would pay for social scheduling.

Lean validation steps: 1. Week 1: Landing page with fake pricing page 2. Week 2: Added email capture when people clicked "Buy" 3. Week 3: Built MVP in 7 days 4. Week 4: First paying customers

Validation metrics: - Landing page: 50% clicked to pricing - Pricing page: 30% clicked "Buy" (showing intent)

Lesson: Test willingness to pay before building.


πŸ“˜ Case Study 3: Zappos

Problem: Would people buy shoes online without trying them?

Lean validation: 1. Founder photographed shoes at local stores 2. Listed them on website 3. When someone ordered, bought shoes and shipped 4. Validated demand before inventory

Lesson: Wizard of Oz MVP validated the riskiest assumption (online shoe buying) with zero inventory risk.


πŸ“˜ Case Study 4: Food on the Table

Problem: Meal planning is hard.

Concierge MVP approach: 1. Found one customer at grocery store 2. Personally planned their meals 3. Shopped with them 4. Created weekly meal plans manually

Scaling: - Added customers one at a time - Understood deep needs - Built technology to automate proven workflows

Lesson: Manual concierge reveals what to automate.


Common Mistakes

❌ Mistakes to Avoid

1. Building Before Validating

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚  THE PREMATURE BUILDING TRAP                                    β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚                                                                 β”‚
β”‚  "I'll just build a quick prototype..." (6 months later)       β”‚
β”‚                                                                 β”‚
β”‚  SYMPTOMS:                                                      β”‚
β”‚  β€’ Feature creep before launch                                  β”‚
β”‚  β€’ Perfectionism on unvalidated ideas                          β”‚
β”‚  β€’ "We're almost ready..."                                      β”‚
β”‚                                                                 β”‚
β”‚  FIX:                                                           β”‚
β”‚  β€’ Set 1-week maximum build sprints                            β”‚
β”‚  β€’ Launch embarrassingly early                                  β”‚
β”‚  β€’ "If you're not embarrassed, you launched too late"          β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

2. Confirmation Bias

The problem: Hearing what you want to hear.

Biased Behavior Correction
Asking friends Ask strangers
Leading questions Open-ended questions
Ignoring negative feedback Seek disconfirming evidence
Small sample size Talk to 20+ people

3. Vanity Metrics Obsession

Signs you're measuring wrong: - Celebrating signups (not usage) - Focusing on traffic (not conversion) - Counting features (not customer outcomes)

4. Pivoting Too Fast/Slow

Too Fast Too Slow
Changing direction after 1 failed experiment Ignoring consistent negative signals
Not giving tests enough time Sunk cost fallacy
Shiny object syndrome "Just one more feature..."

5. Not Talking to Customers

The data tells you what. Customers tell you why.

Aim for: - 5-10 customer conversations before building anything - Weekly customer conversations during building - Direct access to customers (not just through support)


Frequently Asked Questions

❓ How many customer interviews should I do?

Answer: Until you stop learning new things.

Stage Minimum Interviews
Problem discovery 10-20
Solution validation 20-30
Ongoing development 5+/week

You'll know you've done enough when you can predict what the next person will say.


❓ How long should validation take?

Typical timelines:

Validation Type Duration
Quick smoke test 1-2 weeks
Full problem validation 2-4 weeks
Solution validation 4-8 weeks
Business model validation 2-3 months

Remember: Speed matters. Don't let validation become procrastination.


❓ What if customers say they want it but don't buy?

Diagnosis: There's a gap between stated interest and real demand.

Possible Cause Solution
Price too high Test lower price points
Problem not urgent Find more painful version
Not right audience Try different segment
Solution doesn't fit Iterate on approach
Trust lacking Add social proof, reduce risk

❓ Should I validate in my target market if I'm not there?

Answer: Yes, but adjust approach.

If You're Outside Target Market Approach
Different country Video calls, local partners
Different industry Find insider advisors
Different role Shadow real users

Proximity to customers matters for deep understanding.


❓ How do I validate B2B ideas?

B2B validation differences:

Aspect B2C B2B
Sample size 50-100+ 10-20
Interview length 15-30 min 45-60 min
Decision maker Individual Committee
Sales cycle Days Weeks-months
Validation signal Signup LOI, pilot

Best B2B validation: Get a paid pilot or letter of intent.


🎯 Take Action

Ready to validate your idea the lean way?

  1. πŸ“ List your assumptions - Write down what must be true
  2. 🎯 Prioritize riskiest - What kills you if wrong?
  3. πŸ§ͺ Design first experiment - Smallest test possible
  4. πŸƒ Run it this week - Speed matters
  5. πŸ“Š Measure and learn - Be honest about results

Free tool: Quickly check if your niche is already taken with our free niche checker -- no signup required.



Last updated: December 2024