majiayu000 Startup Idea Validation
Systematic validation machine for testing ideas BEFORE building. Validate hypotheses, score opportunities, and make evidence-based GO/NO-GO decisions.
Modern Best Practices (Dec 2025):
- Run a validation ladder (interviews → smoke test → concierge/MVP → paid pilots) before building.
- Pre-register decision thresholds (avoid “moving goalposts” after seeing data).
- Separate evidence quality (strong/medium/weak) from confidence and excitement.
- Test willingness-to-pay early (pricing pages, pilots, LOIs) and treat time-to-value as a constraint.
- Handle customer/market data with purpose limitation, retention, and access controls.
When to Use This Skill
| Trigger | Action |
|---|---|
| "Validate this idea" | Run full 9-dimension validation |
| "Is this worth building?" | Run validation scorecard |
| "Test my hypothesis" | Run hypothesis canvas |
| "Market size for X" | Run market sizing |
| "Should I build X or Y?" | Run comparative validation |
| "What's the riskiest assumption?" | Run RAT analysis |
Validation Ladder (Dec 2025)
| Step | Goal | Strong signal | Weak signal | Output |
|---|---|---|---|---|
| Customer interviews | Validate problem + context | Repeated pain + real workarounds | Hypothetical enthusiasm | Notes + quotes + JTBD |
| Smoke test | Validate demand | Clicks/signups with clear intent | Survey-only interest | Landing page metrics |
| Concierge / Wizard-of-Oz | Validate workflow value | Users complete job and return | One-off curiosity | Learning report |
| Paid pilot | Validate willingness-to-pay | Paid, renewed, or expanded | “Will pay later” | Pilot results + pricing |
Use templates/validation-experiment-planner.md for experiment design and decision thresholds.
9-Dimension Validation Framework
Quick Reference
| Dimension | Weight | Key Question | Score Range |
|---|---|---|---|
| Problem Severity | 15% | Hair on fire or nice to have? | 0-10 |
| Market Size | 12% | Big enough to matter? | 0-10 |
| Market Timing | 10% | Why now? | 0-10 |
| Competitive Moat | 12% | Defensible advantage? | 0-10 |
| Unit Economics | 15% | Can this be profitable? | 0-10 |
| Founder-Market Fit | 8% | Right team for this? | 0-10 |
| Technical Feasibility | 10% | Can we actually build it? | 0-10 |
| GTM Clarity | 10% | Know how to reach customers? | 0-10 |
| Risk Profile | 8% | Manageable risk level? | 0-10 |
Verdict Thresholds
| Score | Verdict | Action |
|---|---|---|
| 80-100 | Strong GO | Proceed to build |
| 60-79 | Conditional GO | Validate riskiest assumptions first |
| 40-59 | PIVOT | Core hypothesis needs rework |
| <40 | NO-GO | Fundamental issues, don't build |
Dimension Deep Dives
1. Problem Severity (15%)
Question: Is this a "hair on fire" problem or a "nice to have"?
Signal Strength Indicators:
| Score | Description | Evidence |
|---|---|---|
| 9-10 | Hair on fire | Customers actively seeking solutions, willing to pay premium |
| 7-8 | Significant pain | Multiple workarounds in use, clear cost of problem |
| 5-6 | Real but manageable | Occasional complaints, spreadsheet solutions exist |
| 3-4 | Nice to have | Would be good but not urgent |
| 1-2 | No real pain | Solution looking for a problem |
Evidence Sources:
- Review mining (G2, Capterra, Reddit) → startup-review-mining
- Customer interviews (5-10 minimum)
- Support ticket analysis
- Search volume for solutions
2. Market Size (12%)
Question: Is this market big enough to build a venture-scale business?
TAM/SAM/SOM Framework:
| Metric | Definition | Minimum Threshold |
|---|---|---|
| TAM | Total Addressable Market | $1B+ |
| SAM | Serviceable Addressable Market | $100M+ |
| SOM | Serviceable Obtainable Market (3yr) | $10M+ |
Sizing Methods:
- Top-down: Industry reports, analyst estimates
- Bottom-up: Customer count × ACV
- Comparable: Similar company revenue extrapolation
3. Market Timing (10%)
Question: Why now? What's changed that makes this possible/necessary?
Timing Signal Matrix:
| Signal | Strong | Weak |
|---|---|---|
| Technology enabler | Just became viable | Has existed for years |
| Regulatory change | New opportunity | No change |
| Behavior shift | COVID/platform shifts changed habits | Status quo |
| Cost curve | 10x cheaper now | Same cost |
| Competition | Market forming | Saturated |
Integration: Use startup-trend-prediction for timing analysis.
4. Competitive Moat (12%)
Question: What will make this defensible over time?
Moat Type Assessment:
| Moat Type | Strength | Build Time | Example |
|---|---|---|---|
| Network Effects | Very Strong | 18-24mo | Marketplace, social |
| Switching Costs | Strong | 12-18mo | Workflow integration |
| Data Moats | Medium | 6-12mo | Proprietary datasets |
| Brand | Medium | 24mo+ | Trust, reputation |
| Regulatory | Strong | Variable | Licenses, compliance |
| Technology | Weak | <6mo | Can be copied |
5. Unit Economics (15%)
Question: Can this be a profitable business?
Key Metrics:
| Metric | Target | Red Flag |
|---|---|---|
| LTV:CAC | >3:1 | <2:1 |
| Payback Period | <12 months | >24 months |
| Gross Margin | >70% (SaaS) | <50% |
| NRR | >100% | <80% |
Formula Basics:
LTV = (ARPU × Gross Margin) / Monthly Churn
CAC = Total Sales & Marketing / New Customers
6. Founder-Market Fit (8%)
Question: Are you the right person/team to solve this?
Assessment Criteria:
| Factor | Strong | Weak |
|---|---|---|
| Domain Expertise | 5+ years in space | No experience |
| Network | Direct access to buyers | Cold outreach only |
| Insight | Unique perspective | Generic understanding |
| Passion | Personal connection | Pure opportunity |
| Ability to Execute | Built similar before | First attempt |
7. Technical Feasibility (10%)
Question: Can we actually build this?
Feasibility Matrix:
| Factor | Score |
|---|---|
| Core technology exists | +3 |
| Similar products exist | +2 |
| Team has built similar | +2 |
| <6 month MVP possible | +2 |
| No regulatory blockers | +1 |
8. GTM Clarity (10%)
Question: Do we know how to reach and convert customers?
GTM Readiness:
| Element | Clear | Unclear |
|---|---|---|
| ICP definition | Specific persona | "Everyone" |
| Acquisition channel | Tested, CAC known | Guessing |
| Sales motion | PLG/Sales/Hybrid decided | TBD |
| Pricing | Market-validated | Assumed |
| First 10 customers | Identified | Unknown |
9. Risk Profile (8%)
Question: What could kill this and how likely?
Risk Categories:
| Risk Type | Example | Mitigation |
|---|---|---|
| Market | Demand doesn't materialize | Validate with pre-sales |
| Technical | Can't build at scale | Prototype early |
| Execution | Team can't deliver | Start small |
| Regulatory | Law changes | Legal review |
| Funding | Can't raise | Bootstrap path |
| Competition | Incumbent pivots | Speed, niche |
Validation Workflow
START
│
▼
┌─────────────────────────────────────┐
│ 1. PROBLEM VALIDATION │
│ - Review mining (10+ sources) │
│ - Customer interviews (5-10) │
│ - Pain severity scoring │
└─────────────────────────────────────┘
│
▼
┌─────────────────────────────────────┐
│ 2. MARKET VALIDATION │
│ - TAM/SAM/SOM calculation │
│ - Timing analysis (trends) │
│ - Competitive landscape │
└─────────────────────────────────────┘
│
▼
┌─────────────────────────────────────┐
│ 3. SOLUTION VALIDATION │
│ - Technical feasibility │
│ - Moat assessment │
│ - Unit economics modeling │
└─────────────────────────────────────┘
│
▼
┌─────────────────────────────────────┐
│ 4. EXECUTION VALIDATION │
│ - Founder-market fit │
│ - GTM clarity │
│ - Risk assessment │
└─────────────────────────────────────┘
│
▼
┌─────────────────────────────────────┐
│ 5. SCORECARD & DECISION │
│ - 9-dimension scoring │
│ - GO/NO-GO verdict │
│ - RAT identification │
└─────────────────────────────────────┘
│
▼
GO / CONDITIONAL GO / PIVOT / NO-GO
Navigation
Resources (Deep Dives)
| Resource | Purpose |
|---|---|
| validation-methodology.md | 9-dimension scoring system details |
| hypothesis-testing-guide.md | How to run validation experiments |
| market-sizing-patterns.md | TAM/SAM/SOM calculation methods |
| moat-assessment-framework.md | Competitive barrier analysis |
Templates (Outputs)
| Template | Use For |
|---|---|
| validation-scorecard.md | Full 9-dimension scoring |
| hypothesis-canvas.md | Hypothesis testing template |
| validation-experiment-planner.md | Hypothesis → method → metric → decision |
| riskiest-assumption-test.md | RAT experiment design |
| market-sizing-worksheet.md | TAM/SAM/SOM calculation |
| go-no-go-decision.md | Final decision template |
Data
| File | Contents |
|---|---|
| sources.json | Validation resources (YC, a16z, SVPG, etc.) |
Integration Points
Receives From
- startup-review-mining - Pain point evidence
- startup-trend-prediction - Market timing data
- startup-competitive-analysis - Competitor data
Feeds Into
- router-startup - Validation results
- product-management - Validated requirements
- startup-business-models - Monetization decisions
Quick Start
Minimum Viable Validation
For rapid first-pass validation:
Pain Check (15 min)
- Search G2/Capterra for competitor complaints
- Search Reddit for problem discussions
- Score: Is this severe enough to trigger action (not “nice to have”)?
Size Check (15 min)
- How many potential customers?
- What would they pay?
- Is the market plausibly large enough for your target outcome? [Inference]
Timing Check (10 min)
- Why now vs 2 years ago?
- What changed?
Competition Check (15 min)
- Who else is doing this?
- What's wrong with existing solutions?
If all 4 checks pass: Proceed to full validation scorecard. If any fail: Pivot or abandon.
Key Principles
Evidence Over Opinion
Every score must have evidence:
- GOOD: "Pain score 8/10: 47 reviews mention this specific complaint (sources linked)"
- BAD: "Pain score 8/10: I think this is a real problem"
Invalidate Fast
Goal is to find reasons NOT to build:
- Cheap to kill ideas, expensive to build
- Seek disconfirming evidence
- Run the riskiest assumption test first
Iterate the Idea, Not Just the Validation
If validation reveals issues:
- Don't just re-score, adjust the idea
- Pivot to stronger position
- Find the version that scores 80+
Do / Avoid (Dec 2025)
Do
- Validate the riskiest assumption first (RAT), not the easiest to test.
- Use the ladder: interviews → smoke → concierge → paid pilots.
- Treat willingness-to-pay as a primary signal, not an afterthought.
- Write decision thresholds before running experiments.
Avoid
- Survey-only validation and hypothetical questions (“Would you use this?”).
- Sampling bias (friends, one subreddit, one review site) without triangulation.
- Building an MVP as “validation” without falsifiable hypotheses.
What Good Looks Like
- ICP: one narrow segment with a clear job, pain severity, and buying trigger.
- Evidence: 10+ direct conversations in the ICP with repeatable pain patterns (not one-off anecdotes).
- WTP: explicit pricing tests and at least one “paid” signal (deposit, pilot fee, LOI with price).
- Experiments: hypotheses + success metrics + stop rules written before execution.
- Decision: a documented go/no-go with the next smallest reversible step.
Optional: AI / Automation
Use only when explicitly requested and policy-compliant.
- Summarization/clustering: speed up synthesis, but keep raw notes + spot-checks.
- Copy drafting: generate landing page variants; humans verify claims and compliance.