Extension: Strategy Deep Dive

“The essence of strategy is choosing what not to do.” - Michael Porter

Why This Extension?

Phase 5A introduced strategy fundamentals. This extension goes deeper, covering advanced frameworks for competitive strategy, value creation, and strategic thinking. Essential for anyone aspiring to C-suite or wanting to think like a business owner.

Prerequisites: Phase 5A (Strategy & Finance)

Week 1-2: What Good Strategy Looks Like

Core Concepts

The Kernel of Good Strategy: A clear diagnosis, a guiding policy, and coherent actions. Most “strategies” fail this basic test.

Bad Strategy Signs: Fluff (buzzwords without substance), failure to face the challenge, mistaking goals for strategy, bad objectives.

Diagnosis: The most important and often skipped step. What is the true nature of the challenge you face?

This Week’s Reading

📖 Good Strategy Bad Strategy by Richard Rumelt (Full book)

• The kernel of good strategy
• Sources of power
• Why bad strategy is so common
• The strategy process

The Kernel of Good Strategy

Bad Strategy vs. Good Strategy

Reflection Questions

1. What is your organization’s current strategy? Does it have a clear kernel?
2. What hard choices is your strategy explicitly making?
3. What challenge is your strategy designed to address?

Week 3-4: Strategic Choices

Core Concepts

The Strategy Choice Cascade: Five interlinked choices that define a winning strategy. Each choice constrains and enables the others.

Where to Play / How to Win: Strategy is fundamentally about these two questions. Many organizations skip “where to play” and jump to capabilities.

Integrated Choice Sets: Strategy isn’t a single decision but a set of reinforcing choices that create competitive advantage.

This Week’s Reading

📖 Playing to Win by A.G. Lafley & Roger Martin (Full book)

• The strategy choice cascade
• Winning aspiration
• Where to play choices
• How to win choices
• Building capabilities

The Strategy Choice Cascade

Where to Play Choices

Application Exercise

Apply the cascade to your organization or a company you admire:

1. What is their winning aspiration?
2. Where do they play (and not play)?
3. How do they win where they play?
4. What capabilities make their strategy possible?

Week 5: Blue Ocean Strategy

Core Concepts

Red Ocean vs. Blue Ocean: Red oceans are existing markets with fierce competition. Blue oceans are new market spaces where competition is irrelevant.

Value Innovation: The simultaneous pursuit of differentiation AND low cost. Breaking the value-cost trade-off.

Four Actions Framework: Eliminate, Reduce, Raise, Create - systematically reconstructing buyer value.

This Week’s Reading

📖 Blue Ocean Strategy by Kim & Mauborgne (Full book)

• Value innovation vs. competition
• Strategy canvas and value curves
• Four actions framework
• Blue ocean examples (Cirque du Soleil, Southwest, etc.)

The Four Actions Framework

Strategy Canvas Example: Cirque du Soleil

Week 6: Strategy Execution & Integration

Core Concepts

The Strategy-Execution Gap: Great strategies fail because of poor execution. Execution requires clear choices cascaded to every level.

Strategic Thinking Habits: The ability to think strategically is a skill that can be developed through practice and reflection.

The Strategist’s Role: Strategy isn’t just for consultants or executives. Everyone can think strategically about their work.

This Week’s Reading

📖 The Strategist by Cynthia Montgomery (Selected chapters)

• The strategist’s distinctive role
• Strategy as continuous process
• Purpose-driven strategy
• Strategy as story

Strategic Thinking Habits

Capstone: Strategic Analysis

Choose an organization and complete a full strategic analysis:

1. Current Strategy: What is their kernel? Is it good or bad strategy?
2. Choice Cascade: Map their five strategic choices
3. Blue Ocean Opportunity: Apply Four Actions Framework
4. Recommendations: What should they change?

Key Frameworks

2026 Update: Strategy in the AI Era

The frameworks above — Rumelt’s kernel, Playing to Win, Blue Ocean — remain the right analytical tools. What has changed is the competitive landscape they operate in. This section applies the frameworks to the AI transition and adds 2026-verified context.

The AI Cost Curve: Fastest Commodity Formation in History

Verified data from Epoch AI (NeurIPS 2025 Workshop):

• The cost to achieve GPT-3.5 level performance dropped over 280-fold between November 2022 and October 2024
• Across benchmarks, the decline rate ranges from 9x to 900x per year, with a median of roughly 50x per year
• GPT-4 launched in 2023 at $30 per million input tokens; GPT-4-equivalent performance is now available for ~$0.10–$0.15 per million tokens

Strategic implication (applying Rumelt): The diagnosis for any company competing in AI is: foundation model capability is commoditizing faster than any previous technology platform. The guiding policy cannot be “build better AI” — the question is where to build genuine advantage on top of AI infrastructure.

Gartner’s framing (cited 2026): Foundation models are now “strategic commodities.” Model performance alone is not a sustainable differentiator.

Build vs. Buy: A TCO Decision, Not an Ideology

McKinsey’s State of AI (November 2025, n=large enterprise sample) identifies four decision factors:

Open-weight models as middle path (2025–2026 trend): Models like Llama 3 and Mistral running on enterprise infrastructure cost 10–12x less than equivalent frontier SaaS. This “build on open-weight” path is growing fastest for companies with some ML capability.

Hidden costs that shift TCO toward buy: data labeling, evaluation dataset creation, legal/compliance review, human escalation design, ongoing monitoring, incident response. These are consistently underestimated in internal build estimates.

Source: McKinsey State of AI 2025; Epoch AI, LLM Inference Price Trends.

AI Competitive Moats: What Is Actually Defensible

Applying the Playing to Win cascade to AI-era strategy:

The data moat caveat: Volume alone is weakening. A static data lake is not a moat. The durable form is a feedback flywheel — a system that learns from ongoing user interaction. Temporal accumulation (decades of unique signals) is irreproducible; bulk scraping is not.

The Productivity Paradox: The Most Important Strategic Fact

McKinsey State of AI, November 2025:

• 88% of organizations use AI in at least one business function
• 72% use generative AI (up from 33% in 2024)
• Only 6% of organizations qualify as “AI high performers” (>5% of EBIT attributable to AI)
• 80%+ of organizations report no measurable bottom-line impact despite widespread adoption

Why the gap? McKinsey’s analysis: organizations are augmenting existing processes rather than redesigning workflows. High performers are 3× more likely to have senior leaders with direct ownership of AI initiatives.

Strategic implication (applying Rumelt’s kernel): The real challenge for most organizations is not “which AI to buy” but “how to redesign work around AI capabilities.” That is a strategy and leadership problem, not a technology problem.

The agentic shift (2025–2026): AI is moving from task-level assistance to workflow automation — agents that plan, use tools, and act across multiple steps. 23% of McKinsey respondents report scaling agentic AI (up from near-zero in 2024). The strategic leaders are those who can redesign workflows for agentic participation, not just add AI to existing roles.

Where Value Is Migrating in the AI Stack

Verified structural trend (multiple analyst sources, 2026):

The value migration pattern resembles historical technology transitions (railroads, internet infrastructure):

• Infrastructure layer (NVIDIA, hyperscalers): capturing value now; NVIDIA data center revenue reached $75.2B in Q1 FY2027
• Applications layer (vertical SaaS, domain-specific tools): the layer where differentiation is building in 2025–2026
• Foundation model layer (OpenAI, Anthropic, Google DeepMind): structural squeeze from both sides — infrastructure costs remain high while application pricing pressure builds

For Blue Ocean strategists: The most underserved competition-free spaces in 2026 are domain-specific AI applications in industries where generic tools cannot match compliance, accuracy, or workflow requirements (healthcare, legal, financial services, industrial).

Note: Foundation model companies’ financial sustainability is genuinely contested — OpenAI’s own projections show losses through 2028. This is relevant context, not settled fact.

Resources

Books

• ⭐⭐⭐ Good Strategy Bad Strategy (Essential - 10h)
• ⭐⭐⭐ Playing to Win (Essential - 7h)
• ⭐⭐⭐ Blue Ocean Strategy (Essential - 9h)
• ⭐⭐ The Strategist (Recommended - 6h)

Free Resources

• Harvard Business Review strategy articles
• McKinsey Quarterly strategy insights
• Blue Ocean Strategy tools - blueoceanstrategy.com

Case Studies

• Apple’s turnaround (1997-2007)
• Southwest Airlines value curve
• Netflix vs. Blockbuster

AI Learning Integration

Strategy Evaluation Prompt

I want to evaluate a company's strategy using Rumelt's framework.

Company: [name or describe the company]

Help me analyze:
1. What is their diagnosis of their challenge?
2. What is their guiding policy?
3. What are their coherent actions?
4. Is this good strategy or bad strategy? Why?

Ask clarifying questions about the company if needed.

Blue Ocean Exploration Prompt

Help me explore blue ocean opportunities for an industry.

Industry: [describe the industry]

Guide me through:
1. Drawing the current strategy canvas (key competing factors)
2. Applying the Four Actions Framework (Eliminate, Reduce, Raise, Create)
3. Identifying a potential blue ocean position
4. Evaluating feasibility and risks

Use questions to help me think through each step.

Phase Assessment

Complete the following to demonstrate strategy competency:

1. Quiz: Strategy Concepts (30%)
2. Case Study: Strategic Analysis (70%)
   • Apply all three major frameworks
   • Evaluate current strategy quality
   • Recommend strategic changes

I'm studying Strategy Deep Dive (Phase EXT-STRATEGY of my MBA program).

Act as a Socratic tutor - don't give me direct answers. Instead, ask me questions to help me discover insights about these concepts: Competitive Strategy, Strategy Execution.

Start by asking what I already know about one of these topics, then guide me deeper with follow-up questions. Challenge my assumptions when appropriate.

After each of my responses, either:
1. Ask a deeper follow-up question
2. Point out a gap in my reasoning
3. Connect my answer to another concept

Let's begin.

Quiz me on Strategy Deep Dive. Ask 10 questions covering: Competitive Strategy, Strategy Execution.

Rules:
- Mix question types (multiple choice, short answer, scenario-based)
- Start easier, get progressively harder
- After each answer, tell me if I'm right or wrong and explain why
- Keep a running score
- At the end, summarize what I know well vs. need to review

Ask the first question now.

Help me apply the main frameworks from this phase to a real situation in my life or work.

First, ask me to describe a recent challenge or decision I faced.

Then guide me through analyzing it using these frameworks:
- Which framework applies best?
- What would each framework reveal about the situation?
- What would I do differently knowing this?

Don't lecture - ask questions that help me discover the insights myself.

I want to practice case analysis for Strategy Deep Dive.

Give me a short business scenario (2-3 paragraphs) involving Competitive Strategy, Strategy Execution.

Then ask me:
1. What's the core problem?
2. Which frameworks from Strategy Deep Dive apply?
3. What biases might cloud judgment here?
4. What would you recommend?

After each answer, push back on my reasoning before moving to the next question.

I'm studying Strategy Deep Dive and need to understand these concepts deeply: Competitive Strategy, Strategy Execution.

For each concept, ask me to explain it in simple terms (as if to a child).

If my explanation is unclear or wrong, don't correct me directly. Instead:
1. Ask clarifying questions
2. Give me a scenario that tests my understanding
3. Help me refine my explanation

The Feynman technique says if you can't explain it simply, you don't understand it well enough.