Recommendation Canvas

Built for PM stakeholder pitches, this helps you fill out a structured canvas that forces clarity on business outcomes, customer outcomes, problem framing, solution hypotheses, positioning, risks, and success metrics. It's adapted from Dean Peters' Productside framework and treats AI features as bets to validate rather than commitments to build. Use it when you need to justify an AI product idea to execs or align cross-functional teams on whether something's worth pursuing. The canvas synthesizes discovery insights you should already have, so don't start here if you haven't done user research yet. It's deliberately comprehensive without being a PRD or full business case.

Install to Claude Code

npx -y skills add deanpeters/product-manager-skills --skill recommendation-canvas --agent claude-code