AI Startup Market Research Framework

Stop Guessing. Start Dominating. - Automated market intelligence for AI startups that refuse to fail.

Go Vertical. Not Ballistic. - Market research and go-to-market strategy development framework for aligning AI startups with investor and market opportunities.

Purpose

Analyze AI startup technical capabilities and strategic positioning against target investor priorities and successful portfolio patterns to identify optimal market opportunities and develop actionable go-to-market strategies.

Source Analysis → Landscape Research → Market Analysis → PMF Analysis → GTM Strategy → Contradiction Analysis → Synthesis → Slide Decks → Pipeline Summary

Usage

Configuration Setup

Before running the pipeline, configure your research targets:

Edit config/sources.md - Add projects/companies to be analyzed and aligned
Edit config/targets.md - Add organizations/markets to align with
Edit config/comments_research.md - Add auxiliary comments, e.g., constraints and focus areas for market research
Edit config/comments_gtm.md - Add auxiliary comments, e.g., specific guidance for go-to-market strategy

Configuration Examples

Adding New Source Projects (config/sources.md):

## Primary Project

- [>>> company name here <<<](https://company.url)
  - Additional information for this source #1
  - Additional information for this source #N

Adding New Alignment Target (config/targets.md):

## Primary Targets

- **Accelerator Program**: https://accelerator.com
  - Program requirements and portfolio analysis

Execution Methods

Method	Command	Description	Use Case
Main Agent Orchestration	`claude -p "Read and execute the workflow described in AGENTS.md"`	Full pipeline with TodoWrite progress tracking, Task tool management, and validation routing	RECOMMENDED - Complete orchestration
Orchestrated with Mode	`claude -p "Execute in [concise\|detailed] mode: Read and execute AGENTS.md"`	Mode-specific execution with automatic subagent transmission	Executive vs research-grade outputs
Legacy Automated	`make orchestrated` / `make all`	Batch execution via Makefile targets	Backward compatibility
Individual Phases	`make landscape research pmf gtm contradictions synthesis slides`	Run specific phases independently	Resume from failure or testing
Development & Debugging	`cat .claude/agents/[subagent].md \| claude -p "execute"`	Direct subagent execution	Phase-specific debugging

Agent-Based Workflow (AGENTS.md)

The new orchestrated workflow in AGENTS.md provides comprehensive pipeline management with the following capabilities:

Key Features

TodoWrite Progress Tracking - Real-time task status updates with pending/in_progress/completed states
Task Tool Subagent Management - Automated subagent execution with proper dependency handling
Parallel Phase Execution - Phase 0 (Source Analysis) and Phase 1A (Industry Landscape) run simultaneously for optimal performance
Validation Loop Integration - Built-in quality assurance with results-validator feedback and correction cycles
Mode Transmission - Automatic detection and transmission of both content depth and strategic approach modes to all subagents
Phase Dependency Management - Ensures proper execution order while optimizing for parallel processing where possible

Execution Flow

Phase 0 + Phase 1A (Parallel) - Source project analysis and industry landscape research execute simultaneously
Phase 1B - Market research analysis using outputs from both Phase 0 and 1A
Phase 2 - Product-market fit analysis based on market research findings
Phase 3 - Go-to-market strategy development leveraging research and PMF insights
Phase 4 - Contradiction analysis across all prior phases
Phase 5 - Research synthesis integrating key findings and resolving contradictions
Phase 6 - Slide deck generation for investor presentations
Phase 7 - Automated pipeline execution summary with strategic insights

Validation Strategy

Default Validation - 1 correction loop (n=1) for each validated phase
Override Options - Skip validation, multiple iterations (n=1 to N), or strict validation (unlimited loops)
Feedback Integration - results-validator feedback automatically forwarded to respective phase subagents for targeted corrections
Quality Assurance - Cross-phase consistency analysis and strategic integration validation

Pipeline Execution Modes

The framework supports dual-mode execution combining content depth and strategic approach:

Content Depth Modes

Mode	Target	Output	Content Depth
Concise	Executive-focused insights	Same file structure (5-6 files per phase)	1-2 pages per file (500-1000 words)
Detailed	Research-grade thoroughness	5-6 comprehensive files per phase	3-5 pages per file (1500-2500 words)

Strategic Approach Modes

Mode	Research Approach	Market Sizing	Growth Strategy	Risk Profile
Conservative	Proven segments, realistic capture rates (0.1-1% TAM)	Lower-end TAM/SOM/SAM estimates	Measured growth (2-5x annually)	Risk-aware, mitigation-focused
Ambitious	Adjacent markets, optimistic capture rates (1-10% TAM)	Higher-end TAM/SOM/SAM estimates	Aggressive growth (5-20x annually)	Opportunity-focused, bold strategies

Mode Combinations (2x2 Matrix):

Concise + Conservative: Executive summaries with safe recommendations and realistic market sizing
Concise + Ambitious: Executive summaries with bold strategies and optimistic projections
Detailed + Conservative: Comprehensive analysis with measured approaches and proven market estimates
Detailed + Ambitious: Comprehensive analysis with aggressive growth plans and expansive market opportunity

Default Modes: Concise + Conservative (executive-focused with risk-aware approach)

Mode Examples:

# Main agent orchestration (RECOMMENDED) - Dual mode combinations
claude -p "Read and execute the workflow described in AGENTS.md"  # Default: concise + conservative
claude -p "Execute in detailed conservative mode: Read and execute AGENTS.md"  # Comprehensive + measured
claude -p "Execute in concise ambitious mode: Read and execute AGENTS.md"      # Executive + bold
claude -p "Execute in detailed ambitious mode: Read and execute AGENTS.md"     # Comprehensive + aggressive

# Legacy automated execution with dual modes
make orchestrated                                    # Default: concise + conservative
make orchestrated DETAIL_MODE=detailed STRATEGY_MODE=conservative  # Comprehensive + measured
make orchestrated DETAIL_MODE=concise STRATEGY_MODE=ambitious      # Executive + bold  
make orchestrated DETAIL_MODE=detailed STRATEGY_MODE=ambitious     # Comprehensive + aggressive

# Individual phases with dual modes
make landscape DETAIL_MODE=detailed STRATEGY_MODE=ambitious
make research pmf gtm DETAIL_MODE=concise STRATEGY_MODE=conservative

Mode Transmission: Main agent automatically detects and transmits both content depth and strategic approach modes to all subagents via Task tool prompts. Both modes logged to results/logs/mode_log.txt.

Pipeline Monitoring

The framework includes logging and progress tracking capabilities.

Pipeline Monitoring - Check current pipeline status:

# View execution logs
cat results/logs/mode_log.txt
ls results/logs/
# Check directory structure and outputs
ls results/

Requirements

Claude Code CLI (npm install -gs @anthropic-ai/claude-code)
Make (build automation)
Node.js and npm (for Claude Code CLI installation)
POSIX Shell (sh, dash, bash - for dashboard and hooks)
Standard Unix Tools (find, wc, awk, date - widely available)

Prerequisites for all methods:

Configuration files setup in config/
Claude Code CLI installed (make setup_claude_code)
Output directories exist (make create_struct)

Automated Features

Auto-creates directory structure via hooks
Real-time progress tracking and notifications
Completion summaries with file listings

For detailed configuration instructions and examples, see individual files in config/ directory and AGENTS.md for orchestration details.

Expected Outcomes

Industry Landscape Research (results/landscape/):

Comprehensive competitive analysis and market mapping
Open-source ecosystem analysis and opportunities
Academic research trends and technical innovations
Technology trends analysis and future directions
Factual industry overview and competitive intelligence

Source Project Analysis (results/source/):

Technical architecture and capability analysis
AI/ML differentiation and competitive advantages
Security, compliance, and scalability assessment
Technical positioning and innovation analysis

Market Research Analysis (results/research/):

Strategic alignment with investor priorities using source analysis
Funding patterns and business model analysis
Investment readiness assessment and recommendations
Market positioning based on technical capabilities

Product-Market Fit Analysis (results/pmf/):

Problem-solution alignment validation
Market readiness assessment
Customer pain point analysis
Competitive positioning evaluation
PMF strength scoring and optimization recommendations

Go-to-Market Strategy (results/gtm/):

Customer segmentation and ideal customer profiles
Value propositions optimized for confirmed product-market fit
Sales channels and acquisition strategies aligned with PMF findings
Launch plan with metrics and timelines leveraging PMF insights

Contradiction Analysis (results/contradictions/):

Cross-phase contradiction detection and resolution frameworks
Gap assessment and vulnerability analysis
Objection handling and counter-argument development
Risk mitigation strategies and conflict resolution

Research Synthesis (results/synthesis/):

Executive summaries highlighting key opportunities
Strategic alignment and implementation priorities
Contradiction resolution strategies and conflict mitigation
Actionable next steps and honest assessments

Slide Deck Presentations (results/slide_decks/):

Investor-ready presentation materials
Visual storytelling with data-driven narratives
Template-based decks (Sequoia, YC, etc.)

Pipeline Execution Summary (results/pipeline_execution_summary.md):

Complete pipeline overview with success metrics
Key strategic insights and alignment analysis
Implementation roadmap and next steps

Project Structure

├── AGENTS.md                # 🎯 MAIN ORCHESTRATION WORKFLOW (RECOMMENDED)
│                            # TodoWrite tracking + Task tool management
│                            # 7-phase pipeline with parallel execution & validation
├── SUBAGENTS.md             # Shared guidelines for all specialized sub-agents
├── .claude/agents/          # Claude Code sub-agent definitions
│   ├── source-project-analyst.md         # Phase 0: Source project technical analysis
│   ├── industry-landscape-researcher.md  # Phase 1A: Industry landscape (parallel with Phase 0)
│   ├── market-research-specialist.md     # Phase 1B: Strategic market research
│   ├── product-market-fit-analyst.md     # Phase 2: Product-market fit validation
│   ├── gtm-strategy-developer.md         # Phase 3: Go-to-market strategy
│   ├── contradiction-analyzer.md         # Phase 4: Contradiction analysis and objection handling
│   ├── research-synthesizer.md           # Phase 5: Cross-analysis and synthesis
│   ├── slide-deck-generator.md           # Phase 6: Presentation materials
│   └── results-validator.md              # Quality assurance and validation (all phases)
├── config/                  # Pipeline configuration and research parameters
│   ├── sources.md           # Projects to be analyzed (Phase 0 input)
│   ├── targets.md           # Organizations/markets to align with
│   ├── comments_research.md # Research constraints and focus areas
│   ├── comments_gtm.md      # GTM strategy guidance
│   └── validation_criteria.md  # Quality standards and validation rules
├── results/                 # Generated pipeline outputs
│   ├── archive/             # Prior runs (timestamped with execution modes)
│   ├── source/              # Phase 0: Source project technical analysis
│   ├── landscape/           # Phase 1A: Industry landscape and competitive intelligence
│   ├── research/            # Phase 1B: Strategic market research and investment analysis
│   ├── pmf/                 # Phase 2: Product-market fit validation
│   ├── gtm/                 # Phase 3: Go-to-market strategy
│   ├── contradictions/      # Phase 4: Contradiction analysis and objection handling
│   ├── synthesis/           # Phase 5: Cross-analysis and strategic integration
│   ├── slide_decks/         # Phase 6: Investor-ready presentations
│   ├── validation/          # Quality assurance reports and feedback summaries
│   └── logs/                # Pipeline execution logs and mode tracking
└── Makefile                 # Legacy pipeline automation (backward compatibility)

Architecture Overview

This framework uses orchestrated sub-agents via the AGENTS.md workflow for automated market research and strategy development:

┌─────────────────────────────────────────────────────────────────────────────┐
│                      MAIN AGENT ORCHESTRATION (AGENTS.md)                   │
│  ┌─────────────────────────────────────────────────────────────────────┐    │
│  │ • TodoWrite Progress Tracking    • Task Tool Subagent Management    │    │
│  │ • Dual Mode Detection & Transmission                                │    |
│  │ • Validation Loop Integration    • Phase Dependency Management      │    │
│  │ • Parallel Execution Optimization • Correction Feedback Routing     │    │
│  └─────────────────────────────────────────────────────────────────────┘    │
└─────────────────┬───────────────────────────────────────────────────────────┘
                  │
                  ▼
┌─────────────────────────────────────────────────────────────────────────────┐
│                      7-PHASE PIPELINE EXECUTION FLOW                        │
│                                                                             │
│ Phase 0: Source Analysis     Phase 1A: Industry Landscape (PARALLEL)        │
│    ↓ results/source/           ↓ results/landscape/                         │
│    └─────────────┬─────────────┘                                            │
│                  ▼                                                          │
│ Phase 1B: Market Research → results/research/  → → → → → → → ┐              │
│                  ↓                                           ↓              │
│ Phase 2: PMF Analysis → results/pmf/                         ↓              │
│                  ↓                                           ↓              │
│ Phase 3: GTM Strategy → results/gtm/  → → → → → → → → → → → →↓              │
│                  ↓                                           ↓              │
│ Phase 4: Contradiction Analysis → results/contradictions/    ↓              │
│                  ↓                                           ↓              │
│ Phase 5: Research Synthesis ← ← ← ← ← ← ← ← ← ← ← ← ← ← ← ← ←┘              │
│          → results/synthesis/                                               │
│                  ↓                                                          │
│ Phase 6: Slide Deck Generation → results/slide_decks/                       │
│                  ↓                                                          │
│ Phase 7: Pipeline Summary → results/pipeline_execution_summary.md           │
│                                                                             │
│ Validation: results-validator → results/validation/[phase]_feedback_*.md    │
│ • Default: 1 correction loop per validated phase                            │
│ • Override: Skip, multiple loops (n=1 to N), or strict (unlimited)          │
└─────────────────────────────────────────────────────────────────────────────┘
                                  ▲
                                  │
┌─────────────────────────────────────────────────────────────────────────────┐
│                           SHARED GUIDELINES & CONFIGURATION                 │
│  ┌──────────────────────────┐    ┌──────────────────────────────────────┐   │
│  │     SUBAGENTS.md         │    │    config/validation_criteria.md     │   │
│  │ • File creation standards│    │ • Quality thresholds & scoring       │   │
│  │ • Citation requirements  │    │ • Source authority hierarchy         │   │
│  │ • Markdown formatting    │    │ • Data freshness requirements        │   │
│  │ • Dual mode adaptation   │    │ • Success criteria definitions       │   │
│  └──────────────────────────┘    └──────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────────────────────┘

Orchestration Layer

Main Agent (AGENTS.md) - Pipeline orchestrator with TodoWrite progress tracking and Task tool subagent management (RECOMMENDED)
- Manages 7-phase execution flow with parallel optimization and dependency tracking
- Handles validation feedback routing with configurable correction loops
- Detects and transmits dual execution modes (concise/detailed + conservative/ambitious) to all subagents
- Provides real-time progress updates and comprehensive pipeline monitoring
Makefile - Legacy batch execution via direct subagent calls for backward compatibility

Sub-Agent Layer (Domain-Specific Execution)

Phase 0: source-project-analyst → results/source/ (analyzes config/sources.md)
Phase 1A: industry-landscape-researcher → results/landscape/
Phase 1B: market-research-specialist → results/research/ (reads source + landscape)
Phase 2: product-market-fit-analyst → results/pmf/ (reads research)
Phase 3: gtm-strategy-developer → results/gtm/ (reads research + PMF)
Phase 4: contradiction-analyzer → results/contradictions/ (reads all prior)
Phase 5: research-synthesizer → results/synthesis/ (reads research + GTM + contradictions)
Phase 6: slide-deck-generator → results/slide_decks/ (reads synthesis)
Validation: results-validator → results/validation/[phase]_*.md (validates any phase)

Shared Guidelines & Configuration

SUBAGENTS.md - Common standards: file creation, citations, markdown formatting (referenced by all subagents)
config/validation_criteria.md - Quality thresholds, source hierarchy, validation standards (referenced by results-validator)

Name		Name	Last commit message	Last commit date
Latest commit History 18 Commits
.claude		.claude
.devcontainer		.devcontainer
assets/images		assets/images
config		config
examples		examples
scripts		scripts
templates/slide_decks		templates/slide_decks
.gitignore		.gitignore
AGENTS.md		AGENTS.md
CHANGELOG.md		CHANGELOG.md
CLAUDE.md		CLAUDE.md
LICENSE		LICENSE
Makefile		Makefile
README.md		README.md
SUBAGENTS.md		SUBAGENTS.md

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

AI Startup Market Research Framework

Purpose

Usage

Configuration Setup

Configuration Examples

Execution Methods

Agent-Based Workflow (AGENTS.md)

Key Features

Execution Flow

Validation Strategy

Pipeline Execution Modes

Content Depth Modes

Strategic Approach Modes

Pipeline Monitoring

Requirements

Automated Features

Expected Outcomes

Project Structure

Architecture Overview

Orchestration Layer

Sub-Agent Layer (Domain-Specific Execution)

Shared Guidelines & Configuration

TODO

About

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Folders and files

Latest commit

History

Repository files navigation

AI Startup Market Research Framework

Purpose

Usage

Configuration Setup

Configuration Examples

Execution Methods

Agent-Based Workflow (AGENTS.md)

Key Features

Execution Flow

Validation Strategy

Pipeline Execution Modes

Content Depth Modes

Strategic Approach Modes

Pipeline Monitoring

Requirements

Automated Features

Expected Outcomes

Project Structure

Architecture Overview

Orchestration Layer

Sub-Agent Layer (Domain-Specific Execution)

Shared Guidelines & Configuration

TODO

About

Resources

License

Code of conduct

Contributing

Security policy

Uh oh!

Stars

Watchers

Forks

Uh oh!

Contributors

Uh oh!

Languages