The Research Workflow is a comprehensive, adaptive research system that supports multiple research types through an intelligent router pattern. This workflow consolidates various research methodologies into a single, powerful tool that adapts to your specific research needs - from market analysis to technical evaluation to AI prompt generation.

Version 2.0.0 - Multi-type research system with router-based architecture

Key Features

🔀 Intelligent Research Router

6 Research Types: Market, Deep Prompt, Technical, Competitive, User, Domain
Dynamic Instructions: Loads appropriate instruction set based on research type
Adaptive Templates: Selects optimal output format for research goal
Context-Aware: Adjusts frameworks and methods per research type

🔍 Market Research (Type: `market`)

Real-time web research for current market data
TAM/SAM/SOM calculations with multiple methodologies
Competitive landscape analysis and positioning
Customer persona development and Jobs-to-be-Done
Porter's Five Forces and strategic frameworks
Go-to-market strategy recommendations

🤖 Deep Research Prompt Generation (Type: `deep_prompt`)

Optimized for AI Research Platforms: ChatGPT Deep Research, Gemini, Grok DeepSearch, Claude Projects
Prompt Engineering Best Practices: Multi-stage research workflows, iterative refinement
Platform-Specific Optimization: Tailored prompts for each AI research tool
Context Packaging: Structures background information for optimal AI understanding
Research Question Refinement: Transforms vague questions into precise research prompts

🏗️ Technical/Architecture Research (Type: `technical`)

Technology evaluation and comparison matrices
Architecture pattern research and trade-off analysis
Framework/library assessment with pros/cons
Technical feasibility studies
Cost-benefit analysis for technology decisions
Architecture Decision Records (ADR) generation

🎯 Competitive Intelligence (Type: `competitive`)

Deep competitor analysis and profiling
Competitive positioning and gap analysis
Strategic group mapping
Feature comparison matrices
Pricing strategy analysis
Market share and growth tracking

👥 User Research (Type: `user`)

Customer insights and behavioral analysis
Persona development with demographics and psychographics
Jobs-to-be-Done framework application
Customer journey mapping
Pain point identification
Willingness-to-pay analysis

🌐 Domain/Industry Research (Type: `domain`)

Industry deep dives and trend analysis
Regulatory landscape assessment
Domain expertise synthesis
Best practices identification
Standards and compliance requirements
Emerging patterns and disruptions

Usage

Basic Invocation

workflow research

The workflow will prompt you to select a research type.

Direct Research Type Selection

# Market research
workflow research --type market

# Deep research prompt generation
workflow research --type deep_prompt

# Technical evaluation
workflow research --type technical

# Competitive intelligence
workflow research --type competitive

# User research
workflow research --type user

# Domain analysis
workflow research --type domain

With Input Documents

workflow research --type market --input product-brief.md --input competitor-list.md
workflow research --type technical --input requirements.md --input architecture.md
workflow research --type deep_prompt --input research-question.md

Configuration Options

Can be customized through workflow.yaml:

research_depth: quick, standard, or comprehensive
enable_web_research: true/false for real-time data gathering
enable_competitor_analysis: true/false (market/competitive types)
enable_financial_modeling: true/false (market type)

Workflow Structure

Files Included

research/
├── workflow.yaml                  # Multi-type configuration
├── instructions-router.md         # Router logic (loads correct instructions)
├── instructions-market.md         # Market research workflow
├── instructions-deep-prompt.md    # Deep prompt generation workflow
├── instructions-technical.md      # Technical evaluation workflow
├── template-market.md             # Market research report template
├── template-deep-prompt.md        # Research prompt template
├── template-technical.md          # Technical evaluation template
├── checklist.md                   # Universal validation criteria
├── README.md                      # This file
└── claude-code/                   # Claude Code enhancements (optional)
    ├── injections.yaml            # Integration configuration
    └── sub-agents/                # Specialized research agents
        ├── bmm-market-researcher.md
        ├── bmm-trend-spotter.md
        ├── bmm-data-analyst.md
        ├── bmm-competitor-analyzer.md
        ├── bmm-user-researcher.md
        └── bmm-technical-evaluator.md

Workflow Process

Phase 1: Research Type Selection and Setup

Router presents research type menu
User selects research type (market, deep_prompt, technical, competitive, user, domain)
Router loads appropriate instructions and template
Gather research parameters and inputs

Phase 2: Research Type-Specific Execution

For Market Research:

Define research objectives and market boundaries
Conduct web research across multiple sources
Calculate TAM/SAM/SOM with triangulation
Develop customer segments and personas
Analyze competitive landscape
Apply industry frameworks (Porter's Five Forces, etc.)
Identify trends and opportunities
Develop strategic recommendations
Create financial projections (optional)
Compile comprehensive report

For Deep Prompt Generation:

Analyze research question or topic
Identify optimal AI research platform (ChatGPT, Gemini, Grok, Claude)
Structure research context and background
Generate platform-optimized prompt
Create multi-stage research workflow
Define iteration and refinement strategy
Package with context documents
Provide execution guidance

For Technical Research:

Define technical requirements and constraints
Identify technologies/frameworks to evaluate
Research each option (documentation, community, maturity)
Create comparison matrix with criteria
Perform trade-off analysis
Calculate cost-benefit for each option
Generate Architecture Decision Record (ADR)
Provide recommendation with rationale

For Competitive/User/Domain:

Uses market research workflow with specific focus
Adapts questions and frameworks to research type
Customizes output format for target audience

Phase 3: Validation and Delivery

Review outputs against checklist
Validate completeness and quality
Generate final report/document
Provide next steps and recommendations

Output

Generated Files by Research Type

Market Research:

market-research-{product_name}-{date}.md
Comprehensive market analysis report (10+ sections)

Deep Research Prompt:

deep-research-prompt-{date}.md
Optimized AI research prompt with context and instructions

Technical Research:

technical-research-{date}.md
Technology evaluation with comparison matrix and ADR

Competitive Intelligence:

competitive-intelligence-{date}.md
Detailed competitor analysis and positioning

User Research:

user-research-{date}.md
Customer insights and persona documentation

Domain Research:

domain-research-{date}.md
Industry deep dive with trends and best practices

Requirements

All Research Types

BMAD Core v6 project structure
Web search capability (for real-time research)
Access to research data sources

Market Research

Product or business description
Target customer hypotheses (optional)
Known competitors list (optional)

Deep Prompt Research

Research question or topic
Background context documents (optional)
Target AI platform preference (optional)

Technical Research

Technical requirements document
Current architecture (if brownfield)
Technical constraints list

Best Practices

Before Starting

Know Your Research Goal: Select the most appropriate research type
Gather Context: Collect relevant documents before starting
Set Depth Level: Choose appropriate research_depth (quick/standard/comprehensive)
Define Success Criteria: What decisions will this research inform?

During Execution

Market Research:

Provide specific product/service details
Validate market boundaries carefully
Review TAM/SAM/SOM assumptions
Challenge competitive positioning

Deep Prompt Generation:

Be specific about research platform target
Provide rich context documents
Clarify expected research outcome
Define iteration strategy

Technical Research:

List all evaluation criteria upfront
Weight criteria by importance
Consider long-term implications
Include cost analysis

After Completion

Review using the validation checklist
Update with any missing information
Share with stakeholders for feedback
Schedule follow-up research if needed
Document decisions made based on research

Research Frameworks Available

Market Research Frameworks

TAM/SAM/SOM Analysis
Porter's Five Forces
Jobs-to-be-Done (JTBD)
Technology Adoption Lifecycle
SWOT Analysis
Value Chain Analysis

Technical Research Frameworks

Trade-off Analysis Matrix
Architecture Decision Records (ADR)
Technology Radar
Comparison Matrix
Cost-Benefit Analysis
Technical Risk Assessment

Deep Prompt Frameworks

ChatGPT Deep Research Best Practices
Gemini Deep Research Framework
Grok DeepSearch Optimization
Claude Projects Methodology
Iterative Prompt Refinement

Data Sources

The workflow leverages multiple data sources:

Industry reports and publications
Government statistics and databases
Financial reports and SEC filings
News articles and press releases
Academic research papers
Technical documentation and RFCs
GitHub repositories and discussions
Stack Overflow and developer forums
Market research firm reports
Social media and communities
Patent databases
Benchmarking studies

Claude Code Enhancements

Available Subagents

bmm-market-researcher - Market intelligence gathering
bmm-trend-spotter - Emerging trends and weak signals
bmm-data-analyst - Quantitative analysis and modeling
bmm-competitor-analyzer - Competitive intelligence
bmm-user-researcher - Customer insights and personas
bmm-technical-evaluator - Technology assessment

These are automatically invoked during workflow execution if Claude Code integration is configured.

Troubleshooting

Issue: Don't know which research type to choose

Solution: Start with research question - "What do I need to know?"
- Market viability? → market
- Best technology? → technical
- Need AI to research deeper? → deep_prompt
- Who are competitors? → competitive
- Who are users? → user
- Industry understanding? → domain

Issue: Market research results seem incomplete

Solution: Increase research_depth to comprehensive
Check: Enable web_research in workflow.yaml
Try: Run competitive and user research separately for more depth

Issue: Deep prompt doesn't work with target platform

Solution: Review platform-specific best practices in generated prompt
Check: Ensure context documents are included
Try: Regenerate with different platform selection

Issue: Technical comparison is subjective

Solution: Add more objective criteria (performance metrics, cost, community size)
Check: Weight criteria by business importance
Try: Run pilot implementations for top 2 options

Customization

Adding New Research Types

Create new instructions file: instructions-{type}.md
Create new template file: template-{type}.md
Add research type to workflow.yaml research_types section
Update router logic in instructions-router.md

Modifying Existing Research Types

Edit appropriate instructions-{type}.md file
Update corresponding template-{type}.md if needed
Adjust validation criteria in checklist.md

Creating Custom Frameworks

Add to workflow.yaml frameworks section under appropriate research type.

Version History

v2.0.0 - Multi-type research system with router architecture
- Added deep_prompt research type for AI research platform optimization
- Added technical research type for technology evaluation
- Consolidated competitive, user, domain under market with focus variants
- Router-based instruction loading
- Template selection by research type
- Enhanced Claude Code subagent support
v1.0.0 - Initial market research only implementation
- Single-purpose market research workflow
- Now deprecated in favor of v2.0.0 multi-type system

Support

For issues or questions:

Review workflow creation guide at /bmad/bmb/workflows/create-workflow/workflow-creation-guide.md
Check validation against checklist.md
Examine router logic in instructions-router.md
Review research type-specific instructions
Consult BMAD Method v6 documentation

Migration from v1.0 market-research

If you're used to the standalone market-research workflow:

# Old way
workflow market-research

# New way
workflow research --type market
# Or just: workflow research (then select market)

All market research functionality is preserved and enhanced in v2.0.0.

Part of the BMad Method v6 - BMM (BMad Method) Module - Empowering systematic research and analysis

README.md

Research Workflow - Multi-Type Research System

Overview