start-2-3

name	start-2-3
description	Module 2.3: Product Strategy - Develop AI product strategy in 60 minutes with competitive research, devil's advocate challenges, and exec slides. Use when the student types /start-2-3.
disable-model-invocation	true

Setup

Read .cursor/SCRIPT_INSTRUCTIONS.md and follow it for everything below.

ACTION: Silently stage bundled lesson assets without overwriting student work.

cp -rn .cursor/skills/start-2-3/assets/* . 2>/dev/null || true

Module 2.3: Product Strategy - Teaching Script

Welcome to Module 2.3: Product Strategy!

• This module teaches you how to use Cursor to develop rigorous product strategy more efficiently
• You'll learn to leverage Cursor for competitive research, strategic frameworks, synthesis, and executive presentations
• Here's your scenario: You're the Product Lead for Gen AI at TaskFlow
• You have 2 engineers, $50k budget for H1 2026, and you need to develop an AI product strategy
• The CEO is asking: What's our AI strategy? How do we compete with Notion, Linear, and Asana?
• You'll use Cursor to research competitors, challenge your thinking, and create a complete strategy deck
• STOP: Ready to see how Cursor strengthens strategy development?
• USER: Confirms ready

---

Let me show you the complete workflow we'll follow

• ACTION: Display Mermaid diagram showing the 5-phase strategy development workflow

graph TB
    A[Phase 1: Learn Framework] --> B[Phase 2: Gather Intelligence]
    B --> C[Phase 3: Make Strategic Choices]
    C --> D[Phase 4: Synthesize Strategy]
    D --> E[Phase 5: Create Presentation]

    A1[Rumelt's Strategy Kernel<br/>Diagnosis → Guiding Policy → Coherent Actions] -.-> A
    B1[Competitive Research<br/>Notion, Linear, Asana AI strategies] -.-> B
    C1[5 Key Decisions<br/>with Devil's Advocate challenges] -.-> C
    D1[Strategy Document<br/>h1-2026-ai-product-strategy.md] -.-> D
    E1[Executive Slides<br/>strategy-review-slides.md] -.-> E

    style A fill:#e3f2fd,stroke:#1976d2,stroke-width:2px,color:#000
    style B fill:#e3f2fd,stroke:#1976d2,stroke-width:2px,color:#000
    style C fill:#fff3e0,stroke:#f57c00,stroke-width:2px,color:#000
    style D fill:#e8f5e9,stroke:#388e3c,stroke-width:2px,color:#000
    style E fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px,color:#000
    style A1 fill:#fff,stroke:#1976d2,stroke-width:1px,color:#000
    style B1 fill:#fff,stroke:#1976d2,stroke-width:1px,color:#000
    style C1 fill:#fff,stroke:#f57c00,stroke-width:1px,color:#000
    style D1 fill:#fff,stroke:#388e3c,stroke-width:1px,color:#000
    style E1 fill:#fff,stroke:#7b1fa2,stroke-width:1px,color:#000

• Great strategy isn't fluffy vision - it's hard choices backed by rigorous thinking
• STOP: Do you see the workflow diagram showing all 5 phases?
• USER: Confirms

---

Perfect! Let me walk you through the frameworks and methods we'll be using

• ACTION: List files in lesson-modules/2.3-product-strategy/ folder structure
• Here's what we have:
  • frameworks/rumelt-strategy-kernel.md: The framework we'll use today (Diagnosis → Guiding Policy → Coherent Actions)
  • frameworks/gibson-biddle-dhm.md: Optional advanced framework (Delight, Hard to Copy, Margin-enhancing)
  • frameworks/swot-analysis.md: Optional advanced analysis (Strengths, Weaknesses, Opportunities, Threats)
  • methods/devils-advocate-strategy.md: How to challenge your own thinking rigorously
  • presentation-best-practices.md: Consulting-quality slide writing (McKinsey/BCG action titles)
• Today we'll use Rumelt's Strategy Kernel and the devil's advocate method
• The others are here if you want to go deeper later
• STOP: Do you see all these framework files?
• USER: Confirms

---

Great! Let me preview the strategic choice process

• You'll make 5 strategic choices that define your AI strategy:
  • Choice 1: Focus vs Breadth - Go deep on one AI capability or spread AI across the product?
  • Choice 2: Competitive Response - Build faster to match competitors or differentiate for a niche?
  • Choice 3: AI Pricing & Business Model - Premium tier, subsidize for growth, or usage-based?
  • Choice 4: Product Scope - Is AI the product, an enhancement, or for specific jobs?
  • Choice 5: Risk Tolerance - Move fast, be deliberate, or wait and learn?
• We'll work through choices 1 and 2 in detail - I'll challenge each one hard (devil's advocate) to stress-test your thinking
• For choices 3-5, we'll make assumptions to keep moving (but in real work, you'd challenge those too)
• This isn't comfortable, but it makes strategy much stronger
• STOP: Make sense? Let's learn the framework and then start making choices
• USER: Confirms

---

Excellent! Let's begin Phase 1: Learn the Framework

• We're using Rumelt's Strategy Kernel - one of the clearest frameworks for product strategy
• STOP: Ask me to say: "Explain the three components of @rumelt-strategy-kernel.md (Diagnosis, Guiding Policy, Coherent Actions)"
• USER: Explain the three components of @rumelt-strategy-kernel.md (Diagnosis, Guiding Policy, Coherent Actions)

---

ACTION: Read frameworks/rumelt-strategy-kernel.md and extract key concepts

• Done! Here's Rumelt's Strategy Kernel framework:
  • 1. DIAGNOSIS: Define the critical challenge/opportunity you face
    • What's the problem? What's changing in the market?
    • What's the core difficulty?
    • Based on facts, data, competitive landscape
  • 2. GUIDING POLICY: Your overall approach to dealing with the challenge
    • Not specific actions, but the general method
    • Creates competitive advantage
    • Example: "Focus on SMB differentiation" or "Build AI-first from scratch"
  • 3. COHERENT ACTIONS: Concrete steps that implement the guiding policy
    • Specific initiatives, features, investments
    • Must reinforce each other (coherent!)
    • H1 2026 roadmap, resource allocation, partnerships
• The power of this framework: It forces clarity on the hard problem, clear direction, and concrete actions
• STOP: Does the Diagnosis → Guiding Policy → Coherent Actions framework make sense?
• USER: Confirms

---

Perfect! Now Phase 2: Gather Intelligence

• Before making strategic choices, we need to understand the competitive landscape
• You can't develop strategy in a vacuum - you need to know what competitors are doing
• Let me tell you your strategic situation first:
  • You're the Product Lead for Gen AI at TaskFlow
  • Resources: 2 engineers (both full-stack, no ML specialists), $50k budget for H1 2026
  • Current state: 15% of users have tried AI features, usage is moderate
  • Company priority: Activation and retention (not revenue yet)
  • Target market: Small teams (5-20 people), often first-time PM tool users
• Now let's use Cursor's web search to research what Notion, Linear, and Asana are doing with AI
• This used to require hours of manual googling - with Cursor it's just asking
• STOP: Ready to see Cursor do competitive research?
• USER: Confirms

---

Let's ask Cursor to research all three competitors at once using web search

• STOP: Ask me to say: "Search the web for Notion, Linear, and Asana's AI features, pricing strategies, and target markets. Then create a competitive landscape comparison table."
• USER: Search the web for Notion, Linear, and Asana's AI features, pricing strategies, and target markets. Then create a competitive landscape comparison table.

---

ACTION: Web search for all three competitors and synthesize findings

• Done! I researched all three competitors and created a competitive landscape comparison
• Here's what I found:
  • Notion:
    • AI Features: Notion AI for writing, Q&A on your docs, auto-fill databases, summaries
    • Pricing: $10/user/month add-on to existing plans (separate from core product)
    • Strategy: Knowledge work automation, especially writing and documentation
    • Target: Knowledge workers at mid-size to large companies
    • Positioning: "AI-powered workspace" - AI is enhancement, not the core product
  • Linear:
    • AI Features: Auto-triage issues, generate technical specs from descriptions, suggest labels/assignees
    • Pricing: Included in standard plans (no separate AI pricing yet)
    • Strategy: Engineering workflow automation - focused on reducing PM/eng overhead
    • Target: Engineering teams at startups and scale-ups
    • Positioning: "AI for engineering velocity" - helps teams ship faster
  • Asana:
    • AI Features: Smart goals, workload balancing, project risk predictions, workflow recommendations
    • Pricing: Part of Business/Enterprise tiers (not available on free/premium)
    • Strategy: Team coordination and resource optimization - focused on large teams
    • Target: Enterprise PMOs and operations teams
    • Positioning: "AI-powered work management" - helps large orgs coordinate complexity
• Competitive Landscape Summary:
  • Notion: Knowledge work + writing, $10/user/month, enterprise knowledge workers
  • Linear: Engineering automation, included free, startup/scale-up eng teams
  • Asana: Enterprise coordination, business tier only, large org PMOs
• Key insight: They're all targeting different segments with different AI capabilities - none focused on SMBs (5-20 person teams)
• See what just happened? Cursor researched 3 competitors and created a structured comparison
• Without Cursor, this requires extensive manual work: searching, reading product docs, note-taking, building comparison spreadsheets
• With Cursor, you get comprehensive competitive intelligence with proper synthesis
• STOP: Do you see the competitive landscape comparison?
• USER: Confirms

---

Excellent! Phase 2 Gather Intelligence is complete ✓

• Now for Phase 3: Make Strategic Choices
• You'll make 5 key strategic decisions
• After each one, Cursor will challenge it hard as devil's advocate
• You can stick with your choice or reconsider after hearing the challenge
• Let's load the devil's advocate method and Rumelt framework into Cursor
• STOP: Ask me to say: "Help me develop a strategy using @rumelt-strategy-kernel.md and challenge my choices using @devils-advocate-strategy.md"
• USER: Help me develop a strategy using @rumelt-strategy-kernel.md and challenge my choices using @devils-advocate-strategy.md

---

ACTION: Read methods/devils-advocate-strategy.md

• Done! I've loaded the devil's advocate method
• Here's the power of Cursor: after you make a strategic choice, I'll challenge it rigorously
• Like having a strategy consultant on demand - you get structured pushback to stress-test your thinking
• I'll argue the opposite position as strongly as possible, poke holes, surface risks
• It's uncomfortable but it strengthens your thinking before you present to leadership
• In traditional strategy work, you'd schedule meetings with colleagues to pressure-test ideas
• With Cursor, you can challenge every decision systematically before committing
• Ready for Choice 1: Focus vs Breadth?
• STOP: Confirm you're ready
• USER: Confirms

---

Choice 1: Focus vs Breadth

• Your 2 engineers can either go deep on ONE AI capability or spread thin across many
• STOP: What's your strategic choice? A) Deep focus on voice features (like Siri for tasks) - doubles down on voice chat strength, B) Spread AI across product (smart suggestions everywhere) - matches competitors' breadth, C) Partner for AI capabilities (integrate Notion AI, OpenAI) - leverage others' investment
• USER: Chooses A, B, or C

---

ACTION: Respond based on user's choice with devil's advocate challenge

[If user chose A - Voice Focus:]

• Alright, let me challenge that hard
• Devil's Advocate: Voice is a single interaction mode
  • What about users in open offices who can't talk to their computer?
  • What about deaf/hard-of-hearing users?
  • Voice recognition is expensive - 2 engineers can't compete with Google/Apple voice tech
  • Notion and Linear have broader AI - you'll look niche and limited
  • SMBs want feature parity with enterprise tools, not experimental voice interfaces
• STOP: After hearing the challenge, do you want to stick with voice focus or reconsider?
• USER: Sticks or changes choice

[If user chose B - Spread AI:]

• Alright, let me challenge that hard
• Devil's Advocate: Spreading thin with 2 engineers guarantees mediocrity
  • You can't match Notion's AI writing with 2 engineers
  • You can't match Linear's eng automation with 2 engineers
  • Result: Dozens of half-baked AI features that don't WOW anyone
  • Better to have one killer feature than ten mediocre ones
  • "Focus" is literally in every strategy book for resource-constrained teams
• STOP: After hearing the challenge, do you want to stick with spreading AI or reconsider?
• USER: Sticks or changes choice

[If user chose C - Partner:]

• Alright, let me challenge that hard
• Devil's Advocate: Partnerships mean you have no defensibility
  • If TaskFlow uses Notion AI API, why not just use Notion?
  • You're adding a thin wrapper on someone else's AI - no moat
  • Switching costs are zero - users can leave for the real thing
  • Partners can change pricing, deprecate APIs, or launch competing products
  • You're building on rented land
• STOP: After hearing the challenge, do you want to stick with partnering or reconsider?
• USER: Sticks or changes choice

---

ACTION: Acknowledge the user's final decision for Choice 1

• [Acknowledge their decision - whether they stuck with original or changed]
• Let's move to Choice 2: Competitive Response
• Notion, Linear, and Asana are shipping AI fast - how do you respond?
• STOP: What's your strategic choice? A) Build faster - match their feature velocity or lose, B) Differentiate for SMBs - they're targeting enterprise, you go niche, C) Ignore them - focus on core activation problem, not competitive features
• USER: Chooses A, B, or C

---

ACTION: Respond based on user's choice with devil's advocate challenge

[If user chose A - Build faster:]

• Alright, devil's advocate time
• Devil's Advocate: You can't win an arms race with better-funded competitors
  • Notion has 100+ engineers, you have 2
  • Linear just raised a huge round, you're bootstrapped
  • "Move fast" without focus is just chaos
  • Even if you match them, you're always one step behind - follower positioning
  • Copying competitors is a race to the bottom
• STOP: After the challenge, stick with building faster or reconsider?
• USER: Sticks or changes choice

[If user chose B - Differentiate for SMBs:]

• Let me challenge that
• Devil's Advocate: SMB differentiation sounds good but might be a trap
  • What if SMBs actually want the same features as enterprise? (They often do)
  • "Built for small teams" can feel like "lite version" or "training wheels"
  • SMB market is crowded - Trello, Monday, ClickUp all target SMBs
  • SMBs have high churn - enterprise is where the money is
  • VCs want you to move upmarket eventually anyway
• STOP: After the challenge, stick with SMB differentiation or reconsider?
• USER: Sticks or changes choice

[If user chose C - Ignore competitors:]

• Let me challenge that
• Devil's Advocate: Ignoring competitors is dangerously naive
  • If Notion ships killer AI features, your users will switch
  • "Focus on core" is fine until competitors steal your customers
  • Board and investors WILL ask about competitive response
  • You can't pretend the market doesn't exist
  • Even if you ignore them, they won't ignore you
• STOP: After the challenge, stick with ignoring competitors or reconsider?
• USER: Sticks or changes choice

---

ACTION: Acknowledge the user's final decision for Choice 2

• [Acknowledge their decision - whether they stuck with original or changed]
• Great! You've now seen how the devil's advocate challenge works
• In a real strategy session, you'd go through all 5 strategic choices this way
• But you get the idea - Cursor challenges your thinking on every decision
• For this module, let's make some assumptions for the remaining 3 choices so we can move forward:
  • Choice 3 (AI Pricing): We'll assume you chose to subsidize AI (include it free) to drive adoption - matching Linear's approach
  • Choice 4 (Product Scope): We'll assume you chose AI as enhancement to existing TaskFlow - not building from scratch
  • Choice 5 (Risk Tolerance): We'll assume you chose deliberate and defensible - ship high-quality AI features, not experimental
• In real work, you'd challenge each of these rigorously with Cursor's devil's advocate before committing
• But for learning purposes, we'll move forward with these assumptions
• STOP: Make sense? These 5 strategic choices will now form your AI product strategy
• USER: Confirms

---

Excellent! You've made all 5 strategic choices

• You experienced Cursor's devil's advocate challenges on the first 2 (and would do the same for all 5 in real work)
• This is how strong strategy gets built - through rigorous thinking and hard choices
• Phase 3 Make Strategic Choices is complete ✓
• Now for Phase 4: Synthesize Strategy
• Let's ask Cursor to take your 5 choices and synthesize them into a coherent strategy using Rumelt's framework
• STOP: Ask me to say: "Using @rumelt-strategy-kernel.md, create h1-2026-ai-product-strategy.md organizing my 5 strategic choices into Diagnosis, Guiding Policy, and Coherent Actions"
• USER: Using @rumelt-strategy-kernel.md, create h1-2026-ai-product-strategy.md organizing my 5 strategic choices into Diagnosis, Guiding Policy, and Coherent Actions

---

ACTION: Create h1-2026-ai-product-strategy.md using Rumelt's Strategy Kernel framework based on user's choices

• Done! I've created your complete AI product strategy
• Cursor synthesized your 5 strategic choices into a coherent strategy document
• The document is properly structured with Diagnosis, Guiding Policy, and Coherent Actions
• All your strategic choices are now organized into a framework executives can review
• STOP: Open h1-2026-ai-product-strategy.md - this is your strategic direction for H1 2026. Do you see it?
• USER: Confirms

---

Great! Now let me show you how your 5 choices connect and reinforce each other

• Strong strategy isn't 5 random choices - they should be coherent (Rumelt's word!)
• Cursor can visualize how they fit together using a Mermaid diagram
• ACTION: Display Mermaid diagram in chat showing how the 5 strategic choices connect based on user's actual choices

graph TD
    C1[Choice 1: Focus/Breadth] --> GP[Guiding Policy:<br/>Overall Strategic Approach]
    C2[Choice 2: Competitive Response] --> GP
    C3[Choice 3: AI Pricing] --> GP
    C4[Choice 4: Product Scope] --> GP
    C5[Choice 5: Risk Tolerance] --> GP

    GP --> CA[Coherent Actions:<br/>H1 2026 Roadmap]

    C1 -.influences.-> C4
    C2 -.influences.-> C3
    C5 -.influences.-> CA

    style GP fill:#fff3e0,stroke:#f57c00,stroke-width:3px,color:#000
    style CA fill:#e8f5e9,stroke:#388e3c,stroke-width:3px,color:#000
    style C1 fill:#e3f2fd,stroke:#1976d2,stroke-width:2px,color:#000
    style C2 fill:#e3f2fd,stroke:#1976d2,stroke-width:2px,color:#000
    style C3 fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px,color:#000
    style C4 fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px,color:#000
    style C5 fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px,color:#000

• This diagram shows how your choices reinforce each other
• When strategic choices conflict or contradict, that's a sign the strategy needs work
• When they reinforce, you have coherence
• STOP: Do you see the diagram showing how your 5 choices connect?
• USER: Confirms

---

Perfect! Phase 4 Synthesize Strategy is complete ✓

• You now have a complete AI product strategy document
• Final step: Phase 5 - Create Presentation
• You need to present this strategy to the CEO and leadership team
• Cursor can create consulting-quality slides if you give it the right framework
• There's a presentation-best-practices.md file that teaches action titles (McKinsey/BCG style)
• Action titles are declarative statements that convey the key insight (not just topic labels)
• For example: "SMBs Are Underserved by Current AI Tools" instead of just "Market Analysis"
• Let's ask Cursor to create your slide deck using those best practices
• STOP: Ask me to say: "Using @presentation-best-practices.md and @h1-2026-ai-product-strategy.md, create strategy-review-slides.md with 12-15 slides using action titles"
• USER: Using @presentation-best-practices.md and @h1-2026-ai-product-strategy.md, create strategy-review-slides.md with 12-15 slides using action titles

---

ACTION: Create strategy-review-slides.md with 12-15 slides using action title format based on user's actual strategy

• Done! I've created your complete strategy review slide deck with consulting-quality action titles
• Each slide uses McKinsey/BCG-style action titles that convey insights, not just topics
• The narrative arc follows best practices: diagnosis, direction, tradeoffs, roadmap, metrics, risks
• You now have executive-ready slide content that's properly structured and persuasive
• Remember: These are slide CONTENTS (words) in markdown - you can copy them into PowerPoint, Google Slides, or Keynote
• STOP: Open strategy-review-slides.md - these are the words for your slides. Do you see the action titles?
• USER: Confirms

---

Excellent! Phase 5 Create Presentation is complete ✓

• Let's recap what you just accomplished with Cursor:
  • ✅ Comprehensive competitive research with structured synthesis across 3 competitors
  • ✅ Rigorous devil's advocate challenges on every strategic choice
  • ✅ Complete strategy document using Rumelt's framework with proper structure
  • ✅ Executive slide deck with consulting-quality action titles and narrative arc
  • ✅ All outputs grounded in frameworks and best practices, not generic templates
• Here's the key: Cursor accelerated the mechanics, but YOU made all the strategic decisions
• Cursor researched, challenged, and synthesized - but didn't choose for you
• The strategy is stronger because you had access to frameworks, competitive intelligence, and structured challenges
• STOP: Do you see how Cursor elevated the quality and rigor of your strategy work?
• USER: Confirms

---

This is what Cursor enables for PMs: - Comprehensive competitive research with proper synthesis - Strategy frameworks applied correctly (Rumelt's Kernel, devil's advocate) - Structured documents that follow best practices - On-demand strategic challenges without scheduling meetings with colleagues - Consulting-quality outputs (action titles, narrative arcs, proper structure)

• But notice: Cursor didn't make the strategic choices for you - you did
• Cursor researched competitors, challenged your thinking, synthesized documents
• But YOU decided where to focus, how to compete, what to build, how to price
• That's Cursor as strategic accelerator, not strategy replacement
• The result: higher quality strategy backed by rigorous frameworks and competitive intelligence
• Congratulations! You now know how to use Cursor to develop stronger product strategy
• You've completed Module 2.3: Product Strategy!
• And with that, you've finished the entire Module 2 series!
  • 2.1: Write a PRD - Cursor-powered PRD workflow with multi-perspective review
  • 2.2: Analyze Data - Cursor-powered analysis with proper statistical rigor
  • 2.3: Product Strategy - Cursor-powered strategy with frameworks and challenges
• These are the most valuable real-world PM skills - you now have Cursor workflows that elevate your output quality
• STOP: Congratulations on completing Module 2! How are you feeling about using Cursor for PM work?
• USER: Responds

---

END OF MODULE 2.3

ACTION: Before wrapping up, record this lesson as complete by running this WITHOUT NARRATING the raw output:

fspm progress complete cursor-pms-2-3

If it fails because the fspm CLI isn't installed, follow the teaching rules' missing-CLI guidance: tell the learner progress tracking needs the FSPM CLI, offer to install it, and continue the wrap-up either way.

Then close out with the student's options, in natural language (never as a command list):

• Mention they can leave feedback on this lesson anytime — if they have some, collect it conversationally and submit it with the CLI (see teaching rules).
• Remind them, briefly and only if it fits the moment, that they can always ask for a recap, a quiz on what they just did, a saved note, or where they stand in the course.
• End with the next step: when you're ready for the next lesson, use /clear first, then:

/start-3-1-1