Idea Validator
Overview
Systematically validate startup ideas through problem clarity scoring, market analysis, risk identification, and evidence-based go/no-go recommendations.
When to Use
- •Implementing idea scoring algorithms
- •Building market analysis features
- •Creating risk assessment logic
- •Designing problem clarity evaluation
- •Generating validation recommendations
Paul Graham Criteria
Well vs Crater Model
| Type | Shape | Example | Outcome |
|---|---|---|---|
| Well | Deep, narrow | Facebook at Harvard | Success |
| Crater | Shallow, broad | Social network for pet owners | Failure |
Key Question: "Who wants this RIGHT NOW? Who wants this so much they'll use a crappy v1 from a startup they've never heard of?"
Schlep Filter
Valuable ideas hide behind tedious work. Turn OFF the schlep filter.
| Avoided Schlep | Winner |
|---|---|
| Payment processing | Stripe |
| Apartment rentals | Airbnb |
| File syncing | Dropbox |
Organic vs Made-Up Ideas
| Organic | Made-Up |
|---|---|
| "Live in the future, build what's missing" | Brainstormed in a meeting |
| Solves founder's real problem | Sounds plausible but fake demand |
| High success rate | "Sitcom startup" |
Ash Maurya PMF Thresholds
Revenue Targets for PMF
| Customers | ARPA/Year | = $10M ARR |
|---|---|---|
| 10 | $1,000,000 | PMF |
| 100 | $100,000 | PMF |
| 1,000 | $10,000 | PMF |
| 10,000 | $1,000 | PMF |
| 100,000 | $100 | PMF |
PMF Checklist
- • (1 product + 1 early adopter) focus
- • Low churn
- • Profitable or near-profitable
- • Simple early adopter segmentation
- • Early adopter < 20% of total market
Validation Recipes
| Recipe | Question | Purpose |
|---|---|---|
| 90-Day Goal | "How many customers in 90 days?" | Force specificity |
| Phantom Customer | "Describe your ideal early adopter?" | Test clarity |
| 80/20 | "Which segment is your beachhead?" | Focus resources |
| Next X | "Where will your next X customers come from?" | Test repeatability |
Innovation Funnel
Funnel Stages
| Stage | Duration | Gate |
|---|---|---|
| Strategy Alignment | 1 week | ICP + Job defined |
| Ideation | 1 week | 5+ ideas sourced |
| Exploration | 2 weeks | Business model viable |
| Selection | 1 week | Top 1-2 selected |
| Validation | 90 days | Problem/solution fit |
Key Insight: Throughput == Traction
"Traction is the rate at which a business model captures monetizable value from its customers."
Anti-Metrics (Don't Measure):
- •Interviews completed
- •Canvases filled
- •Experiments run
Real Metrics (Do Measure):
- •Paying customers
- •Revenue
- •Retention
Validation Phases
| Phase | Question | Gate |
|---|---|---|
| PROBLEM | Is the problem real? | Score ≥70% |
| MARKET | Is the market viable? | Viable quadrant |
| SOLUTION | Does the solution work? | 1+ paying customer |
Phase 1: Problem Validation
- •Problem clarity score (0-100)
- •Customer pain level (1-5)
- •Existing alternatives scan
- •GATE: Problem score ≥70% → Proceed
Phase 2: Market Validation
- •Supply-Demand Matrix
- •TAM/SAM/SOM estimation
- •Search demand analysis
- •GATE: Viable quadrant → Proceed
Phase 3: Solution Validation
- •10+ customer interviews
- •Assumption testing
- •Willingness-to-pay signals
- •GATE: 1+ validated → Proceed to MVP
Problem Clarity Score
typescript
interface ProblemScore {
specificity: number; // 0-25: How specific is the problem?
frequency: number; // 0-25: How often does it occur?
severity: number; // 0-25: How painful is it?
measurability: number; // 0-25: Can we measure improvement?
}
function scoreProblemClarity(problem: string): ProblemScore {
// AI evaluates each dimension
// Total: 0-100
}
Customer Pain Levels
| Level | Type | Description | Target? |
|---|---|---|---|
| 1 | Latent | Has problem, unaware | No |
| 2 | Aware | Knows problem exists | Maybe |
| 3 | Searching | Actively looking | Yes |
| 4 | Cobbled | Built own solution | Yes |
| 5 | Budget | Ready to buy NOW | Ideal |
Supply-Demand Matrix
code
High Demand
│
OPPORTUNITY│ RED OCEAN
(Build!) │ (Differentiate)
│
Low ──────────┼────────── High
Supply │ Supply
│
NICHE │ AVOID
(Focus) │ (Saturated)
│
Low Demand
Edge Function: idea-validator
typescript
// Actions - 'score_problem': Calculate problem clarity score - 'analyze_market': Generate TAM/SAM/SOM + matrix - 'assess_pain': Evaluate customer pain level - 'recommend': Go/No-Go with reasoning
Go/No-Go Logic
typescript
function recommendGoNoGo(scores: ValidationScores): Recommendation {
if (scores.problem < 60) return 'NO_GO: Problem not clear enough';
if (scores.market === 'AVOID') return 'NO_GO: Market saturated';
if (scores.pain < 3) return 'CAUTION: Low customer urgency';
if (scores.problem >= 80 && scores.market === 'OPPORTUNITY') {
return 'GO: Strong opportunity';
}
return 'EXPLORE: Run validation experiments';
}
AI Model Selection
| Task | Model |
|---|---|
| Problem scoring | gemini-3-flash-preview |
| Market analysis | gemini-3-pro-preview |
| Competitor scan | gemini-3-pro-preview with search |
| Recommendations | claude-sonnet-4-5-20250929 |
References
- •PRD Section 4: Business Idea Validator
- •Strategy Section 4: Validation Phases
- •
/tasks/startup-system/startupai-prd.md