Measure Experiment Results

After you run an A/B test or experiment, this helps you document what actually happened with proper statistical rigor. It walks you through presenting primary and secondary metrics, analyzing results by segment, extracting learnings beyond the numbers, and making a clear ship/iterate/kill recommendation. The quality checklist pushes you to include confidence intervals and honestly report negative results, which is refreshing. Most teams are terrible at turning individual experiments into organizational knowledge, so having a structured way to capture not just outcomes but learnings and next steps is legitimately useful for building institutional memory.

Install to Claude Code

npx -y skills add product-on-purpose/pm-skills --skill measure-experiment-results --agent claude-code