product-discovery

プロダクトディスカバリー

基本原則

• 継続的なディスカバリー — 単発の調査ではなく、毎週ユーザーと対話します
• 成果主導 — 構築するソリューションではなく、達成すべき成果から始めます
• 仮説の検証 — リソースを投入する前に、リスクの高い仮説を検証します
• 共創 — 顧客のためだけでなく、顧客と一緒に構築します
• データ主導 — 直感やステークホルダーの意見ではなく、根拠を活用します
• 問題優先 — ソリューションを考案する前に、問題領域を深く理解します

---

厳守事項（必ず守ること）

以下のルールは必須です。違反すると、このスキルは正しく機能しません。

ソリューションありきで考えない

決してソリューションから始めてはいけません。必ず最初に問題と成果を定義します。

❌ FORBIDDEN:
"We should build a search bar for the product page"
"Let's add AI recommendations"
"Users need a mobile app"

✅ REQUIRED:
"Problem: Users can't find products (40% exit rate on catalog)
Outcome: Reduce exit rate to 20%
Possible solutions:
1. Search bar with filters
2. AI-powered recommendations
3. Better category navigation
4. Visual product browsing"

根拠に基づく意思決定

実際のユーザー調査による根拠なしに、ユーザーのニーズを決めつけてはいけません。

❌ FORBIDDEN:
- "Users probably want X" (assumption without data)
- "Our competitor has X, so we need it too" (copycat without validation)
- "The CEO thinks we should build X" (HiPPO without evidence)
- "It's obvious users need X" (intuition without validation)

✅ REQUIRED:
- "5 out of 8 interviewed users mentioned X as a pain point"
- "Analytics show 60% of users abandon at step 3"
- "Prototype test: 7/10 users completed task successfully"
- "Survey (n=500): 45% rated feature as 'must have'"

インタビューの最低基準

セグメントごとに5人以上のユーザーにインタビューせずに、問題を検証してはいけません。

❌ FORBIDDEN:
- "We talked to 2 users and they loved the idea"
- "One customer requested this feature"
- "Based on a quick chat with sales..."

✅ REQUIRED:
| Segment | Interviews | Key Finding |
|---------|------------|-------------|
| Power Users | 6 | 5/6 struggle with X |
| New Users | 5 | 4/5 drop off at onboarding |
| Churned | 5 | 3/5 cited missing feature Y |

Minimum per segment: 5 interviews
Confidence increases with more interviews

反証可能な仮説

すべての仮説は、明確な成功基準を設けて検証可能かつ反証可能にする必要があります。

❌ FORBIDDEN:
- "Users will like the new design" (not falsifiable)
- "This will improve engagement" (no success criteria)
- "The feature will be useful" (vague)

✅ REQUIRED:
| Assumption | Test | Success Criteria | Result |
|------------|------|------------------|--------|
| Users will complete onboarding in new flow | Prototype test with 10 users | >70% completion | TBD |
| Users prefer visual search | A/B test | >10% lift in conversions | TBD |
| Price point is acceptable | Landing page test | >3% conversion | TBD |

---

クイックリファレンス

どの場面で何を使うか

シナリオ	フレームワーク／ツール	成果物
プロダクトのアイデアを検証する	プロダクト機会評価	Go／No-Go の判断
市場機会の規模を測る	TAM／SAM／SOM	市場規模の推定値
ユーザーニーズを理解する	ユーザー調査（インタビュー、アンケート）	ユーザーのインサイト、ペインポイント
競合を分析する	競合分析	競合環境マップ
ユーザーの動機を発見する	Jobs-to-be-Done（JTBD）	ジョブストーリー、成果
機能に優先順位を付ける	カノモデル	機能の分類
価値提案を定義する	バリュープロポジションキャンバス	価値提案の文言
プロダクトのコンセプトをテストする	リーンスタートアップ／MVP	検証済みの学び
機会をマッピングする	オポチュニティ・ソリューション・ツリー	優先順位付けされた機会

---

継続的なディスカバリーの習慣

プロダクトトリオ

ディスカバリーは、3つの役割が毎週協力して主導します。

Product Manager → Defines outcomes, owns roadmap
Designer        → Explores solutions, tests usability
Engineer        → Assesses feasibility, proposes technical solutions

毎週の活動

## 1. Customer Interviews (Weekly)
- Schedule 3-5 interviews per week minimum
- Mix of current users, churned users, prospects
- Focus on understanding problems, not pitching solutions
- Record and share insights with team

## 2. Assumption Testing (Weekly)
- Identify riskiest assumptions about solutions
- Design quick tests (prototypes, landing pages, fake doors)
- Run experiments with real users
- Measure results against success criteria

## 3. Opportunity Mapping (Ongoing)
- Build opportunity solution tree
- Map customer needs to potential solutions
- Prioritize based on impact and feasibility
- Update as you learn

ディスカバリーとデリバリー

Discovery (What to Build)          Delivery (How to Build It)
├─ Customer interviews             ├─ Sprint planning
├─ Prototype testing               ├─ Development
├─ Assumption validation           ├─ QA testing
├─ Market research                 ├─ Deployment
└─ Opportunity assessment          └─ Post-launch monitoring

Key difference: Discovery reduces risk BEFORE committing to build

---

プロダクト機会評価

Marty Caganの10の質問

プロダクトの取り組みを始める前に、次の質問に答えます。

## 1. Problem Definition
**What problem are we solving?**
- Be specific and measurable
- Validate it's a real problem (not assumed)

## 2. Target Market
**For whom are we solving this problem?**
- Define specific user segments
- Size the addressable market (TAM/SAM/SOM)

## 3. Opportunity Size
**How big is the opportunity?**
- Revenue potential
- User growth potential
- Strategic value

## 4. Success Metrics
**How will we measure success?**
- Leading indicators (usage, engagement)
- Lagging indicators (revenue, retention)
- Define targets upfront

## 5. Alternative Solutions
**What alternatives exist today?**
- Direct competitors
- Indirect solutions
- Current user workarounds

## 6. Our Advantage
**Why are we best suited to solve this?**
- Unique capabilities
- Market position
- Technical advantages

## 7. Strategic Fit
**Why now? Why us?**
- Market timing
- Strategic alignment
- Resource availability

## 8. Dependencies
**What do we need to succeed?**
- Technical dependencies
- Partnership requirements
- Regulatory considerations

## 9. Risks
**What could go wrong?**
- Market risk (will anyone want it?)
- Execution risk (can we build it?)
- Monetization risk (will they pay?)

## 10. Cost of Delay
**What happens if we don't build this?**
- Competitive disadvantage
- Lost revenue
- Market opportunity window

価値と労力のフレームワーク

機会にすばやく優先順位を付けます。

High Value, Low Effort  → Do First (Quick Wins)
High Value, High Effort → Plan Strategically (Big Bets)
Low Value, Low Effort   → Do Later (Fill Gaps)
Low Value, High Effort  → Don't Do (Money Pit)

---

ディスカバリーの手法

どの場面でどの手法を使うか

## Generative Research (What problems exist?)
Use when: Starting new product area, exploring unknown space
Methods:
- Ethnographic field studies
- Contextual inquiry
- Diary studies
- Open-ended interviews

## Evaluative Research (Does our solution work?)
Use when: Testing specific solutions, validating designs
Methods:
- Usability testing
- Prototype testing
- A/B testing
- Concept testing

## Quantitative Research (How much? How many?)
Use when: Need statistical validation, measuring impact
Methods:
- Surveys
- Analytics analysis
- A/B experiments
- Market sizing

## Qualitative Research (Why? How?)
Use when: Understanding motivations, uncovering insights
Methods:
- User interviews
- Focus groups
- Customer advisory boards
- User observation

インタビューのベストプラクティス

## Preparation
- Define research goals and hypotheses
- Create interview guide (but stay flexible)
- Recruit right participants (6-8 per segment)
- Schedule 45-60 min sessions

## During Interview
✓ Ask open-ended questions ("Tell me about...")
✓ Follow up with "Why?" 5 times to get to root cause
✓ Listen more than talk (80/20 rule)
✓ Ask about past behavior, not future hypotheticals
✓ Look for workarounds and pain points
✓ Record and take notes

✗ Don't ask leading questions
✗ Don't pitch your solution
✗ Don't ask "Would you use X?" (people lie)
✗ Don't multi-task while interviewing

## Example Questions
- "Walk me through the last time you [did task]"
- "What's most frustrating about [current solution]?"
- "How are you solving this problem today?"
- "What would make [task] easier for you?"
- "Tell me more about that..."

アンケートのベストプラクティス

## When to Survey
✓ Validate findings from qualitative research
✓ Measure satisfaction or sentiment at scale
✓ Prioritize features (Kano surveys)
✓ Segment users by behavior/needs

## Survey Design
- Keep it short (<10 min to complete)
- One question per screen on mobile
- Mix question types (multiple choice, scale, open-ended)
- Avoid leading or biased questions
- Test survey with 5 people before sending

## Question Types
- Multiple choice → Segmentation, categorization
- Likert scale (1-5) → Satisfaction, importance
- Open-ended → Qualitative insights
- Ranking → Prioritization
- NPS (0-10) → Loyalty measurement

## Distribution
- In-app surveys (high response, biased to engaged users)
- Email surveys (broader reach, lower response)
- Incentivize thoughtful responses ($10 gift card, early access)
- Follow up with interviews for interesting responses

---

2025年のプロダクトディスカバリーのトレンド

AIを活用した調査

## AI Tools for Discovery
- **Insight synthesis** — AI analyzes interview transcripts, identifies patterns
- **Synthetic personas** — AI-generated user proxies for rapid testing
- **Market intelligence** — AI tracks competitor moves, pricing changes
- **Survey analysis** — Automated sentiment analysis, theme extraction
- **Trend detection** — AI identifies emerging market trends early

## Examples
- Crayon → Competitive intelligence automation
- Glimpse → Trend detection from web data
- Delve AI → Automated persona creation
- Attest → AI-powered survey insights
- Quantilope → Machine learning research automation

## Best Practices
✓ Use AI to scale research, not replace human insight
✓ Validate AI findings with real user conversations
✓ Combine AI analysis with qualitative depth
✗ Don't rely solely on synthetic users
✗ Don't skip talking to real customers

大規模な継続的ディスカバリー

## Modern Approach
- Discovery is embedded in every sprint, not a phase
- Weekly user touchpoints (interviews, tests, feedback)
- Rapid experimentation (dozens of tests running)
- Fast pivots based on evidence (days, not months)

## Team Structure
- Product trios own discovery for their area
- Centralized research team supports (tools, methods)
- Customer success shares feedback loop
- Data analysts provide quantitative insights

## Cadence
- Weekly: Customer interviews, prototype tests
- Bi-weekly: Opportunity review, assumption validation
- Monthly: Market analysis, competitive review
- Quarterly: Strategic discovery (new markets, big bets)

---

オポチュニティ・ソリューション・ツリー

概要

成果からソリューションまでの道筋をマッピングするビジュアルフレームワークです。

        OUTCOME (Business goal)
             |
    ┌────────┴────────┐
    │                 │
OPPORTUNITY 1    OPPORTUNITY 2
    │                 │
    ├─ Solution A     ├─ Solution C
    ├─ Solution B     └─ Solution D
    └─ Solution C

作成方法

## Step 1: Define Outcome
Start with measurable business outcome
Example: "Increase Day 30 retention from 20% to 30%"

## Step 2: Map Opportunities
Discover customer needs/pain points through research
Example: "Users don't understand core features"

## Step 3: Generate Solutions
For each opportunity, brainstorm multiple solutions
Example:
- Better onboarding tutorial
- In-app tooltips
- Interactive product tour

## Step 4: Test Assumptions
For each solution, identify riskiest assumption and test
Example: "Users will complete a 5-step tutorial"
Test: Build simple prototype, test with 10 users

## Step 5: Compare Solutions
Use evidence to choose best path forward
Build what tests validate, discard what fails

メリット

✓ Visualizes multiple paths to outcome
✓ Prevents jumping to first solution
✓ Encourages broad exploration before narrowing
✓ Documents why decisions were made
✓ Keeps team aligned on priorities

---

参考情報（詳細）

## Integrating Discovery with Delivery から始まる詳細な資料は、このスキルを簡潔に保つため、reference/extended.md に移動しました。移動された例、コマンドカタログ、チェックリスト、プラットフォームの詳細、または実装テンプレートが必要なタスクでは、この参考資料を読み込んでください。