LogoSkills

/research — 근거 있는 조사 보고서 만들기

시장·경쟁사·기술·사용자를 대신 조사해 출처 링크가 붙은 질문별 답변, 경쟁사/가격 비교표, 핵심 인사이트, 기간별(2주/1~3개월/3개월+) 권고를 담은 `research-*.md` 한 편을 냅니다.

/research — 근거 있는 조사 보고서 만들기#

항목내용
실행 명령/cc-product:research

한마디로#

"이거 한번 알아봐 줘"라고 맡기면, 시장·경쟁사·기술·사용자를 대신 조사해서 근거(출처 링크)와 실행 권고가 담긴 한 편의 보고서로 정리해 주는 명령입니다. 도서관 사서와 컨설턴트를 합쳐 놓은 역할이라고 보면 됩니다.

누가·언제 쓰나요#

  • "피트니스 앱 시장이 얼마나 큰지", "경쟁사 대비 우리 강점이 뭔지" 같은 막연한 궁금증을 정리된 답으로 받고 싶을 때
  • 제품 기획·아키텍처 설계 전, 결정의 근거가 될 자료가 필요할 때
  • 네 가지 조사 유형 중 하나를 골라서 진행합니다: 시장 / 경쟁사 / 기술 / 사용자 (혹은 이들의 혼합)
  • 보통 30~90분 정도 걸리는 작업입니다.

무엇을 해주나요#

조사가 끝나면 research-(주제)-(날짜).md 파일 한 개가 만들어집니다. 그 안에는:

  • 핵심 요약(Executive Summary) — 바쁜 사람이 먼저 읽을 3줄 요약
  • 질문별 답변 — 처음에 정한 질문들에 대해 "근거+확신도(상/중/하)"까지 붙인 답
  • 경쟁사 비교표 / 가격 비교표 — 경쟁사 조사일 때 한눈에 보이는 표
  • 핵심 인사이트 5~10개 — 단순 정보가 아니라 "그래서 우리는 무엇을 해야 하나"
  • 권고 사항 — 당장(2주) / 단기(1~3개월) / 장기(3개월+)로 나눈 실행 계획
  • 출처 목록 — 모든 주장에 붙는 링크

어떻게 쓰나요#

# 조사 워크플로우 시작
/research

명령을 실행하면 다음을 차례로 물어봅니다. 답만 하면 됩니다.

  • 무엇을 조사할지 (예: "프로젝트 관리 도구 시장의 경쟁사")
  • 조사 유형 (시장 / 경쟁사 / 기술 / 사용자 / 혼합 중 택1)
  • 꼭 답을 듣고 싶은 질문 3~7개
  • 범위·제약 (예: "미국 시장만", "B2B SaaS만", "React 생태계만")

안에서 무슨 일이 벌어지나요#

조사를 8단계로 쪼개서 차근차근 진행합니다.

  1. 범위 정하기 — 위 질문들로 무엇을·왜·어디까지 조사할지 확정합니다.
  2. 방법 고르기 — 조사 유형에 맞춰 웹 검색·문서 수집 등 어떤 방법을 쓸지 정합니다.
  3. 자료 수집 — 실제로 검색하고 자료를 모읍니다 (경쟁사는 3~7곳, 기술은 후보별로).
  4. 분석 — 질문마다 답을 정리하고, 출처가 서로 어긋나는 부분이나 빈틈을 짚어냅니다.
  5. 비교표 작성 — 경쟁사·가격을 표로 정리합니다 (해당하는 경우만).
  6. 인사이트 도출 — "그래서 어떻게 해야 하나"를 우선순위와 함께 5~10개 뽑습니다.
  7. 보고서 작성 — 위 내용을 하나의 문서로 묶어 파일로 저장합니다.
  8. 상태 갱신·다음 단계 안내 — 조사 유형에 맞는 다음 작업(기획·설계 등)을 추천합니다.

⚙️ 상세 옵션·실행 명세 (개발자 / AI 에이전트용)

Workflow Overview#

Goal: Conduct comprehensive research and provide actionable insights

Phase: Cross-phase (supports all BMAD phases)

Agent: Creative Intelligence

Inputs: Research topic, research type, specific questions

Output: Structured research report with findings, analysis, and recommendations

Duration: 30-90 minutes


Pre-Flight#

  1. Load context per helpers.md#Combined-Config-Load
  2. Explain purpose:

    "I'll conduct comprehensive research on your topic. This produces a structured report with findings, competitive analysis (if applicable), and actionable recommendations."


Research Process#

Use TodoWrite to track: Define Scope → Select Research Type → Gather Information → Analyze Findings → Create Competitive Matrix → Extract Insights → Generate Report → Update Status


Part 1: Define Research Scope#

Ask user:

Q1: Research Topic

"What are we researching?"

Examples:

  • Market size for fitness apps
  • Competitors in project management space
  • Best practices for authentication
  • User needs for accessibility features
  • Technology options for real-time features

Store as: {{research_topic}}

Q2: Research Type

"What type of research?"

Options:

  1. Market Research - Market size, trends, growth, customer segments
  2. Competitive Research - Competitors, features, positioning, gaps
  3. Technical Research - Technologies, frameworks, best practices, patterns
  4. User Research - User needs, pain points, behaviors, journeys
  5. Mixed - Combination of above

Store as: {{research_type}}

Q3: Specific Questions

"What specific questions should this research answer?"

List 3-7 key questions to guide research.

Store as: {{research_questions}}

Q4: Constraints

"Any constraints or focus areas?"

Examples:

  • Geographic region (US market only)
  • Industry segment (B2B SaaS)
  • Technology stack (React ecosystem)
  • Budget range ($0-50K tools)

Store as: {{constraints}}


Part 2: Select Research Methods#

Based on research type, determine methods:

For Market Research:

  • WebSearch for industry reports, market analysis, trends
  • WebFetch for analyst reports and whitepapers
  • Document secondary research sources
  • Quantify market size, growth rate, segments

For Competitive Research:

  • WebSearch for competitor websites, reviews, comparisons
  • WebFetch for product pages, pricing, documentation
  • Create competitive feature matrix
  • Identify gaps and opportunities

For Technical Research:

  • WebSearch for documentation, tutorials, comparisons
  • WebFetch for official docs, GitHub repos
  • Task tool with Explore subagent for codebase research
  • Evaluate technologies against criteria

For User Research:

  • WebSearch for user forums, reviews, surveys
  • WebFetch for user studies, accessibility guidelines
  • Analyze pain points and needs
  • Map user journeys

Inform user:

"Research approach:

  • Method 1: {{method}}
  • Method 2: {{method}}
  • Method 3: {{method}}

Starting research..."


Part 3: Gather Information#

Execute research using appropriate tools.

Market Research

Use WebSearch for:

-  " {{market}} market size 2025 " 
 -  " {{market}} industry trends " 
 -  " {{market}} growth projections " 
 -  " {{market}} customer segments "

Capture:

  • Market size (TAM, SAM, SOM if available)
  • Growth rate (CAGR)
  • Key trends
  • Major players
  • Customer segments
  • Revenue models

Competitive Research

For each competitor (3-7 competitors):

Use WebSearch:

-  " {{competitor_name}} features " 
 -  " {{competitor_name}} pricing " 
 -  " {{competitor_name}} reviews " 
 -  " {{competitor_name}} vs alternatives "

Use WebFetch on:

  • Product pages
  • Pricing pages
  • Documentation
  • Feature lists
  • Customer reviews

Capture per competitor:

### {{Competitor Name}}

**Overview:** {{description}}
**Target Market:** {{target}}
**Pricing:** {{pricing_model}}
**Key Features:**
- Feature 1
- Feature 2
- Feature 3

**Strengths:**
- Strength 1
- Strength 2

**Weaknesses:**
- Weakness 1
- Weakness 2

**Unique Differentiators:** {{what_makes_them_unique}}

**Source:** {{url}}

Technical Research

For each technology/framework:

Use WebSearch:

-  " {{technology}} documentation " 
 -  " {{technology}} best practices " 
 -  " {{technology}} vs {{alternative}} " 
 -  " {{technology}} performance benchmarks "

Use WebFetch for:

  • Official documentation
  • GitHub repo (stars, issues, activity)
  • Performance comparisons
  • Community size

If researching internal codebase:

  • Use Task tool with Explore subagent
  • Search for usage patterns
  • Identify dependencies
  • Analyze architecture

Capture per technology:

### {{Technology Name}}

**Purpose:** {{what_it_does}}
**Maturity:** {{stable/beta/experimental}}
**Community:** {{size_indicators}}
**Performance:** {{benchmarks}}

**Pros:**
- Pro 1
- Pro 2

**Cons:**
- Con 1
- Con 2

**Best For:** {{use_cases}}
**Avoid If:** {{anti_patterns}}

**Source:** {{url}}

User Research

Use WebSearch for:

-  " {{user_type}} pain points {{domain}} " 
 -  " {{user_type}} needs {{domain}} " 
 -  " user reviews {{related_products}} " 
 -  " accessibility requirements {{domain}} "

Use WebFetch for:

  • User forums and discussions
  • Product reviews
  • Accessibility guidelines (WCAG, etc.)
  • User research reports

Capture:

  • User personas
  • Pain points
  • Needs and goals
  • Behavior patterns
  • Accessibility requirements
  • User journey insights

Part 4: Analyze Findings#

Synthesize all gathered information.

For each research question from Part 1:

### Q: {{research_question}}

**Answer:** {{synthesis_from_research}}

**Supporting Evidence:**
- {{source_1}}: {{finding}}
- {{source_2}}: {{finding}}
- {{source_3}}: {{finding}}

**Confidence:** High | Medium | Low
**Gaps:** {{what_we_still_dont_know}}

Identify patterns:

  • Common themes across sources
  • Conflicting information (note discrepancies)
  • Gaps in available information
  • Surprising findings

Part 5: Create Competitive Matrix (if applicable)#

If research type is Competitive or Mixed, create feature comparison.

Matrix format:

## Competitive Feature Matrix

| Feature | Our Product | Competitor 1 | Competitor 2 | Competitor 3 |
|---------|-------------|--------------|--------------|--------------|
| Feature 1 |Planned ||||
| Feature 2 |||||
| Feature 3 |Unique ||||

Legend:
-= Available
-= Not available
-Planned = On roadmap
-Unique = Our differentiator

Pricing comparison (if applicable):

## Pricing Comparison

| Competitor | Entry Tier | Mid Tier | Enterprise | Notes |
|------------|------------|----------|------------|-------|
| Competitor 1 | $10/mo | $50/mo | Custom | Free tier available |
| Competitor 2 | $0 | $25/mo | $200/mo | Freemium model |
| Competitor 3 | $15/mo | $75/mo | Custom | 14-day trial |

Part 6: Extract Key Insights#

Identify 5-10 actionable insights from research.

Format each insight:

### Insight {{N}}: {{title}}

**Finding:** {{what_research_revealed}}

**Implication:** {{what_this_means_for_our_project}}

**Recommendation:** {{what_we_should_do}}

**Priority:** High | Medium | Low

**Supporting Data:** {{sources_and_specifics}}

Categorize insights:

  • Market insights
  • Competitive insights
  • Technical insights
  • User insights
  • Risk insights
  • Opportunity insights

Part 7: Generate Research Report#

Create research report per helpers.md#Apply-Variables-to-Template

Use template: research-report.md (or generate inline)

Report structure:

# Research Report: {{research_topic}}

**Date:** {{date}}
**Research Type:** {{research_type}}
**Duration:** {{duration}}

## Executive Summary

{{2-3_paragraph_summary}}

Key findings:
- Finding 1
- Finding 2
- Finding 3

## Research Questions

{{questions_from_part_1}}

## Methodology

**Research approach:**
- {{method_1}}
- {{method_2}}
- {{method_3}}

**Sources:** {{count}} sources consulted

**Time period:** {{date_range_of_research}}

## Findings

### Research Question 1: {{question}}
{{answer_from_part_4}}

### Research Question 2: {{question}}
{{answer_from_part_4}}

[All questions...]

## Detailed Analysis

{{If market research:}}
### Market Overview
- Market Size: {{size}}
- Growth Rate: {{rate}}
- Key Trends: {{trends}}
- Customer Segments: {{segments}}

{{If competitive research:}}
### Competitive Landscape
{{competitive_matrix_from_part_5}}

#### Competitor Profiles
{{detailed_competitor_analysis_from_part_3}}

#### Competitive Gaps
- Gap 1: {{what_competitors_lack}}
- Gap 2: {{opportunity_for_differentiation}}

{{If technical research:}}
### Technology Evaluation
{{technology_comparisons_from_part_3}}

#### Recommended Technology Stack
- Technology 1: {{rationale}}
- Technology 2: {{rationale}}

{{If user research:}}
### User Insights
- Pain Points: {{findings}}
- Needs: {{findings}}
- Behavior Patterns: {{findings}}
- Accessibility: {{requirements}}

## Key Insights

{{insights_from_part_6}}

## Recommendations

### Immediate Actions (Next 2 weeks)
1. {{action_1}}
2. {{action_2}}

### Short-term (Next 1-3 months)
1. {{action_1}}
2. {{action_2}}

### Long-term (3+ months)
1. {{action_1}}
2. {{action_2}}

## Research Gaps

**What we still don ' t know:**
- Gap 1: {{unanswered_question}}
- Gap 2: {{area_needing_deeper_research}}

**Recommended follow-up research:**
- Follow-up 1
- Follow-up 2

## Sources

1. {{source_1}} - {{url}}
2. {{source_2}} - {{url}}
3. {{source_3}} - {{url}}

[All sources...]

## Appendix

{{Any additional data, charts, or detailed comparisons}}

---

*Generated by BMAD Method v6 - Creative Intelligence*
*Research Duration: {{duration}} minutes*
*Sources Consulted: {{count}}*

Save to: {{output_folder}}/research-{{topic}}-{{date}}.md

Inform user:

Research Complete!

Research Type: {{type}}
Sources Consulted: {{count}}
Key Insights: {{count}}

Document: {{file_path}}

Top 3 Insights:
1. {{insight_1_title}}
2. {{insight_2_title}}
3. {{insight_3_title}}

Top Recommendation: {{top_recommendation}}

Update Status#

Per helpers.md#Update-Workflow-Status

Update bmm-workflow-status.yaml:

last_workflow: research
last_workflow_date: {{current_date}}
research:
  reports_completed: {{increment_count}}
  last_research_topic: {{research_topic}}
  last_research_type: {{research_type}}
  total_sources: {{total_count}}

Recommend Next Steps#

Based on research type and findings:

If Market Research → Business Analyst or Product Manager

Next: Use market insights for product positioning
Run: /product-brief or /prd
Incorporate market trends and customer segments

If Competitive Research → Product Manager

Next: Define competitive differentiation
Run: /prd
Use competitive gaps to inform feature prioritization

If Technical Research → System Architect

Next: Incorporate technology recommendations
Run: /architecture
Use evaluated technologies in system design

If User Research → Product Manager or UX Designer

Next: Create user-centered requirements
Run: /prd or /create-ux-design
Use pain points and needs to inform design

If Research revealed gaps → Creative Intelligence

Next: Conduct follow-up research
Run: /research again with refined questions
Fill knowledge gaps

If Research supports hypothesis → Continue to planning

Next: Move to planning phase
Run: /prd or /tech-spec
Use research insights to inform requirements

Helper References#

  • Load config: helpers.md#Combined-Config-Load
  • Apply template: helpers.md#Apply-Variables-to-Template
  • Save document: helpers.md#Save-Output-Document
  • Update status: helpers.md#Update-Workflow-Status
  • Determine next: helpers.md#Determine-Next-Workflow

Notes for LLMs#

  • Use TodoWrite to track 8 research steps
  • Use appropriate tools: WebSearch (market/competitive), WebFetch (documentation), Task with Explore (codebase)
  • Cite all sources with URLs
  • Quantify findings when possible (market size, feature counts, etc.)
  • Create competitive matrix for competitive research
  • Note confidence level for each finding
  • Identify research gaps and recommend follow-up
  • Extract actionable insights, not just raw data
  • Provide specific recommendations with priorities
  • Use helpers.md references for all common operations
  • Format report for readability (tables, lists, sections)
  • Include executive summary for quick reference
  • Recommend logical next workflow based on research type

Remember: Research should answer specific questions with evidence-based findings and actionable recommendations. Always cite sources and quantify when possible.