Writing Product Strategy

Overview

Creates a multi-year (3-5 year) product strategy document. Defines vision, strategic pillars, capability roadmap, and competitive moats. Connects daily tactical work to long-term strategic direction.

When to Use

Annual/bi-annual strategy planning
New product area or market entry
Major pivot or repositioning
Investor/board presentations
Connecting quarterly charters to long-term vision

Core Pattern

Step 1: Define Scope & Horizon

Ask user:

"What product area is this strategy for?"
"What time horizon? (3 years or 5 years)"
"What existing context should I read?" (charters, VOC, competitive analysis)

Step 2: Vision Statement

Craft 2-3 sentences answering:

Where should we be in N years?
What will be different about the world/market?
What role will our product play?

Good vision example: "In 5 years, every retail buyer will use our Business Network as their primary source of truth for catalog data. Suppliers will publish once to reach all retailers. We'll be the operating system for retail product information."

Bad vision example: "Be the best catalog platform" (vague, no specific outcome)

Step 3: Market Context

Analyze multi-year trends:

Trends: What's changing in retail/CPG/catalogs over next N years?
- Technology shifts (AI, automation, standards)
- Buyer behavior changes
- Regulatory changes
- Industry consolidation
Opportunities: What white space can we capture?
- Underserved segments
- Emerging use cases
- Platform expansion
Threats: What could disrupt us?
- Competitive moves
- Technology disruption
- Market consolidation
- Commoditization

Step 4: Strategic Pillars

Define 3-5 strategic pillars. Each pillar:

Why this matters: Business case for investing here
Key capabilities: What we need to build
Success looks like: Outcome in N years

Example pillar: "Pillar: Automated Catalog Quality Why: 40% of product data is incorrect, causing returns/support burden Capabilities: AI validation, auto-enrichment, supplier feedback loops Success: <5% data error rate, trusted source of truth"

Step 5: Capability Roadmap

Map major capabilities to timeline:

Year 1: [Foundational capabilities]
Year 2: [Building on Year 1]
Year 3: [Advanced capabilities]
Dependencies: [What needs to exist first]

Step 6: Competitive Moats

Define what makes this defensible:

Network effects: Gets better with more users?
Data moat: Proprietary data/insights?
Switching costs: Hard for customers to leave?
Brand/trust: Reputation/relationships?
Technology: Unique technical advantage?

Step 7: Strategy Cascade

Show how strategy connects to execution:

Strategy → Charters: How quarterly charters advance pillars
Charters → PRDs: How PRDs deliver capabilities
Metrics: Strategy KPIs → Charter metrics → PRD acceptance criteria

Step 8: Strategic Reasoning (Meta-Prompt)

Before generating the main strategy document, create a "Strategic Reasoning" section following .claude/rules/pm-core/meta-prompt-reasoning.md:

Problem & Goals: Restate what success looks like for this strategy
Context & Constraints: List available evidence and gaps
Options Evaluated: What strategic directions were considered? (at least 3)
Selection Rationale: Why these pillars beat alternatives

This transparency helps stakeholders verify assumptions and challenge logic before committing to a multi-year direction.

Step 9: Generate Output

Write to outputs/strategy/product-strategy-YYYY.md:

markdown

---
generated: YYYY-MM-DD HH:MM
skill: writing-product-strategy
horizon: [3-year / 5-year]
sources:
  - outputs/roadmap/Q1-2026-charters.md (modified: YYYY-MM-DD)
  - outputs/insights/voc-synthesis-latest.md (modified: YYYY-MM-DD)
  - (market research from [source])
downstream:
  - outputs/roadmap/Qx-YYYY-charters.md (should align with pillars)
---

# Product Strategy: [Product Area] ([Start Year]-[End Year])

## Strategic Reasoning

### Problem & Goals
**Problem:** [One-sentence restatement of strategic challenge]
**Success Criteria:** [What makes this strategy high-quality]

### Context & Constraints
**Available Evidence:**
- [Source 1 with key findings]
- [Source 2 with key findings]

**Constraints:**
- [Resource/market/technical limitations]

**Gaps:**
- [What we don't know that matters]

### Strategic Options Evaluated

| Option | Strengths | Weaknesses | Evidence Score |
|--------|-----------|------------|----------------|
| [Direction A] | [Pros] | [Cons] | High/Medium/Low |
| [Direction B] | [Pros] | [Cons] | High/Medium/Low |
| [Direction C] | [Pros] | [Cons] | High/Medium/Low |

### Selection Rationale
**Chosen:** [Strategic pillars selected]

**Why chosen beats alternatives:**
1. [Reason 1 with evidence]
2. [Reason 2 with evidence]
3. [Reason 3 with evidence]

**What would change this strategy:**
- [Market/competitive condition that would shift direction]

---

## Vision Statement

[2-3 sentences: Where we'll be in N years, what will be different, our role]

## Market Context

### Trends (Next [N] Years)

| Trend | Impact on Our Market | Our Response |
|-------|---------------------|--------------|
| [Technology trend] | [How it changes things] | [How we'll adapt/capitalize] |
| [Buyer behavior trend] | [How it changes things] | [How we'll adapt/capitalize] |
| [Industry trend] | [How it changes things] | [How we'll adapt/capitalize] |

### Opportunities (White Space)

| Opportunity | Market Size | Why Now | Why Us |
|-------------|-------------|---------|--------|
| [Underserved segment] | [TAM if known] | [Market timing] | [Our advantage] |
| [Emerging use case] | [TAM if known] | [Market timing] | [Our advantage] |

### Threats (Risks to Vision)

| Threat | Impact if Realized | Mitigation |
|--------|-------------------|------------|
| [Competitive threat] | High/Med/Low | [How we'll counter] |
| [Technology disruption] | High/Med/Low | [How we'll adapt] |
| [Market shift] | High/Med/Low | [How we'll pivot] |

## Strategic Pillars

### Pillar 1: [Name]

**Why This Matters:**
[Business case - what problem does this solve? What value does it create?]

**Key Capabilities:**
- [Capability 1]: [What it enables]
- [Capability 2]: [What it enables]
- [Capability 3]: [What it enables]

**Success Looks Like (Year [N]):**
- [Outcome metric 1]: [Target]
- [Outcome metric 2]: [Target]
- [Outcome metric 3]: [Target]

### Pillar 2: [Name]

**Why This Matters:**
[Business case]

**Key Capabilities:**
- [Capability 1]: [What it enables]
- [Capability 2]: [What it enables]
- [Capability 3]: [What it enables]

**Success Looks Like (Year [N]):**
- [Outcome metric 1]: [Target]
- [Outcome metric 2]: [Target]
- [Outcome metric 3]: [Target]

### Pillar 3: [Name]

**Why This Matters:**
[Business case]

**Key Capabilities:**
- [Capability 1]: [What it enables]
- [Capability 2]: [What it enables]
- [Capability 3]: [What it enables]

**Success Looks Like (Year [N]):**
- [Outcome metric 1]: [Target]
- [Outcome metric 2]: [Target]
- [Outcome metric 3]: [Target]

## Capability Roadmap

| Capability | Year 1 | Year 2 | Year 3 | Dependencies | Pillar |
|------------|--------|--------|--------|--------------|--------|
| [Cap 1] | [Milestone] | [Milestone] | [Milestone] | [What needs to exist first] | Pillar 1 |
| [Cap 2] | [Milestone] | [Milestone] | [Milestone] | [What needs to exist first] | Pillar 2 |
| [Cap 3] | [Milestone] | [Milestone] | [Milestone] | [What needs to exist first] | Pillar 1 |

**Key:**
- Year 1: [Current year] - [Current year]
- Year 2: [Current year + 1]
- Year 3: [Current year + 2]

## Competitive Moats

### Moat 1: [Type - Network Effect / Data / Switching Cost / Brand]
**What:** [Description of the moat]
**How We Build It:** [Actions to strengthen]
**Defensibility:** [Why competitors can't replicate]

### Moat 2: [Type]
**What:** [Description]
**How We Build It:** [Actions to strengthen]
**Defensibility:** [Why hard to replicate]

### Moat 3: [Type]
**What:** [Description]
**How We Build It:** [Actions to strengthen]
**Defensibility:** [Why hard to replicate]

## Strategy Cascade to Execution

### How Strategy Drives Quarterly Work

Strategy (3-5 years) ↓ (Advance strategic pillars) Quarterly Charters (This quarter) ↓ (Deliver capabilities) PRDs (This sprint) ↓ (Build features) Engineering Execution


**Example:**
- **Strategy Pillar:** Automated Catalog Quality
- **Q1 Charter:** Ship AI validation for top 5 error types
- **PRD:** AI validation service for missing attributes
- **Success Metric Cascade:**
  - Strategy KPI: <5% data error rate (Year 3)
  - Charter Metric: Reduce error rate from 40% to 30% (Q1)
  - PRD Acceptance: Validate 10K products, catch 90% of missing attributes

### Alignment Check

Every quarterly charter should answer:
- **Which pillar does this advance?** [Pillar name]
- **Which capabilities does this build?** [Capability from roadmap]
- **How does this move us toward vision?** [Connection to vision]

If a charter doesn't advance a pillar → question whether it belongs on roadmap.

## Investment Allocation (Target)

| Pillar | % of Engineering | Rationale |
|--------|-----------------|-----------|
| Pillar 1 | [%] | [Why this allocation] |
| Pillar 2 | [%] | [Why this allocation] |
| Pillar 3 | [%] | [Why this allocation] |
| KTLO/Tech Debt | [%] | [Maintenance baseline] |
| **Total** | 100% | |

## Key Assumptions

| Assumption | Impact if Wrong | Validation Plan |
|------------|----------------|-----------------|
| [Market grows at X%] | [Impact] | [How we'll validate] |
| [Customers will pay for Y] | [Impact] | [How we'll validate] |
| [Technology Z will be viable] | [Impact] | [How we'll validate] |

## Unknowns / Open Questions

- [What market data is missing?]
- [What customer validation needed?]
- [What technical feasibility unclear?]

## Sources Used
- [file paths]
- [market research sources]
- [customer interview data]

## Claims Ledger

| Claim | Type | Source |
|-------|------|--------|
| [Market trend X] | Evidence | [Analyst report or article] |
| [Customer pain Y] | Evidence | [VOC:line or interviews] |
| [Technology feasibility Z] | Assumption | [Needs tech spike] |
| [Market size $A] | Evidence/Unknown | [Source or "Need research"] |

Step 10: Copy to History & Update Tracker

Run pm-os mirror --quiet to copy to history/writing-product-strategy/product-strategy-YYYY-MM-DD.md
Update nexa/state.json and append to outputs/audit/auto-run-log.md

Step 11: Post-Skill Reflection (MANDATORY)

Follow protocol in .claude/rules/pm-core/post-skill-reflection.md:

Extract key learnings (3-5 insights):
- What strategic options were evaluated vs selected?
- How did meta-prompt reasoning help clarify choices?
- What evidence gaps existed?
- What pillars emerged as most defensible?
- Connections to past strategies or charters?
Create learning entry:
- Write to history/learnings/YYYY-MM-DD-writing-product-strategy.md
- Use template from post-skill-reflection rule
Create insight beads:
- For each significant, reusable insight
- Append to .beads/insights.jsonl
- Types: insight (strategic learning), pattern (recurring), question (validation needed)

Request output rating:

Rate this strategy doc (1-5, or 'skip'):
1 - Needs major revision
2 - Below expectations
3 - Meets expectations
4 - Exceeds expectations
5 - Outstanding, exactly what I needed

If rated: Create output-rating bead
Capture any qualitative feedback

Detect decisions:
- Strategy creation = high confidence decision
- Auto-log: "Chose strategic pillars: [list] based on [evidence]"
- Write to outputs/decisions/YYYY-MM-DD-product-strategy-[horizon].md

Report completion:

✅ Product strategy complete → outputs/strategy/product-strategy-YYYY.md
   Mirrored to history/writing-product-strategy/product-strategy-YYYY-MM-DD.md

📝 Captured learnings: [N] insights, [N] beads → history/learnings/YYYY-MM-DD-writing-product-strategy.md
📋 Logged decision: [Horizon] product strategy → outputs/decisions/

Rate this strategy doc (1-5, or 'skip'): [prompt for rating]

Quick Reference

Good vs Bad Strategic Pillars

Bad Pillar	Why Bad	Good Pillar
"Improve quality"	Vague, no specific outcome	"Automated catalog quality: <5% error rate via AI validation"
"Faster performance"	Generic	"Sub-second search: 500ms p95 for 100M SKUs enabling real-time buyer workflows"
"Better UX"	Unmeasurable	"Zero-training onboarding: 80% task completion without docs/support"

Common Strategy Mistakes

Too tactical: Listing features, not capabilities/outcomes
No moats: Strategy can be copied easily
Disconnected from reality: Vision ignores current state/constraints
No cascade: Strategy doesn't connect to quarterly charters
Assumes linear growth: Ignores market shifts/disruption
Missing evidence: Based on assumptions, not customer/market data

Verification Checklist

Vision statement clear (2-3 sentences)
Market context documented (trends, opportunities, threats)
3-5 strategic pillars defined
Each pillar has why/capabilities/success criteria
Capability roadmap maps to years
Competitive moats identified
Strategy cascade to execution shown
Investment allocation defined
Key assumptions documented
Metadata header complete
Copied to history, tracker updated

Evidence Tracking

Claim	Type	Source
[Market trend]	Evidence	[Research source]
[Customer need]	Evidence	[VOC or interviews]
[Competitive position]	Evidence/Assumption	[Analysis or assessment]
[Technology feasibility]	Assumption	[Needs validation]

Search AI Tools

writing-product-strategy

Install this agent skill to your Project

SKILL.md