hypothesis-tree_skill

Overview

This skill structures complex questions into clear, testable hypotheses using a hypothesis tree approach. It helps convert vague problems into MECE (Mutually Exclusive, Collectively Exhaustive) branches so teams can prioritize and run focused tests. Use it to drive faster, evidence-based decisions for product, research, and debugging scenarios.

How this skill works

It frames a central question, generates first-level hypotheses grouped to be MECE, and decomposes each branch into directly testable sub-hypotheses. Each hypothesis is evaluated for evidence, impact, and test effort to produce a prioritized testing plan. The tree is updated as results arrive, guiding follow-up analysis or experiments.

When to use it

• Validating new product or feature ideas
• Investigating unexpected metric changes (e.g., drop in retention)
• Planning user research or experiments
• Breaking down ambiguous strategic or technical problems
• Prioritizing what to test first and aligning stakeholders

Best practices

• Start with a specific, measurable central question
• Ensure branches are mutually exclusive and collectively exhaustive (MECE)
• Decompose until hypotheses are falsifiable and actionable
• Prioritize tests by impact, effort, and existing evidence
• Update the tree iteratively as new data arrives
• Share the tree visually for stakeholder alignment

Example use cases

• Why is signup conversion under target? — test Awareness, Ability, Motivation, Technical branches
• Investigate sudden churn increase — test Product changes, Market shifts, Customer mix, Service issues
• Plan an experiment roadmap for a new feature by mapping adoption obstacles and quick wins
• Frame user research topics by turning ambiguous feedback into measurable hypotheses
• Communicate analysis structure before a cross-functional investigation or demo