The Claude Strategy Playbook for AI PMs with Shubham Sharan

Most PMs use Claude like a chatbot. The best use it like a strategy team.

Claude can write documents, summarise research and generate ideas. But its real value for product managers goes much further.

You can use it to pressure-test a board narrative, size a market using sources you can defend, analyse competitors, build a business case and estimate the cost of an AI feature before finance asks.

This playbook shows you how to do that, step by step.

Part 1: Learn Claude End to End

This section covers the Claude features a senior PM should know and how to use them in real work.

1. Choose the right model for each task: Haiku, Sonnet, Opus or Fable
2. Understand the cost and trade-offs of each model
3. Learn eight prompting techniques from Anthropic’s courses
4. Use a master prompt template for complex product and strategy work
5. Set up Projects, Artifacts, Research, Memory, Skills, Connectors and MCP
6. Use Cowork, scheduled tasks and Claude in Chrome
7. Use Claude Code even if you do not write code
8. Understand the difference between AI workflows and agents
9. Learn five common agent patterns
10. Create evals to measure the quality of AI features
11. Work through an example that reduces cost per request by up to 90%

Part 2: Run 14 Strategy Plays

This section turns proven consulting frameworks into practical Claude workflows.

Each play explains when to use it, what steps to follow, which prompt to copy, what the final output should include and which mistakes to avoid.

Frame

Situation Assessment

Decision Question

MECE Issue Tree

Hypothesis Design

Analyse

Market Sizing

Competitive Teardown

Customer Segmentation

Profit Pool and Value Proposition

Decide

Strategic Options and Trade-offs

Business Case

Build, Buy or Partner

Communicate

Roadmap and First 90 Days

Risk Register

Pyramid Story and Board Narrative

Part 3: Put It Into Practice

The final section helps you start using Claude in your day-to-day work.

1. A 30-day implementation plan
2. A team adoption checklist
3. A one-page guide to AI governance
4. A printable cheat sheet