data-define

Define business problems, structure issue trees, and design KPIs in analysis_context.md.

Updated Mar 3, 2026
One-click install
npx skills add https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace --skill data-define
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: data-define
Source: https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace/tree/main/plugins/data-analysis/skills/data-define
Command: npx skills add https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace --skill data-define

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps teams convert vague business goals into a structured analysis plan by defining the central question, building an issue tree, listing testable hypotheses, and specifying measurable KGI/KPI targets so analysis leads to actionable decisions.

Core Features & Use Cases

  • Issue Framing: Guide users to write a single, actionable Central Question and check for actionability and data feasibility.
  • Issue Tree & Hypotheses: Create a MECE issue tree and enumerate hypotheses with rationale, falsifying evidence, and validation approaches.
  • KPI Design & Sample Sizing: Define KGI/KPI with quantitative success criteria and provide basic sample-size calculation examples and guidance.
  • Use Case: A product analyst preparing an A/B test can use this Skill to formalize the objective, set KPIs with MDE, estimate required sample sizes, and produce Next Steps for data collection and EDA.

Quick Start

Use data-define to frame the central question, produce an issue tree, list hypotheses, and update analysis_context.md with KPIs and next steps.

Frequently Asked Questions about data-define

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I define KPIs and hypotheses from ambiguous business goals?

To define KPIs and hypotheses from ambiguous business goals, you frame a single actionable Central Question, build a MECE issue tree, and enumerate testable hypotheses with falsifying evidence and validation approaches.

What is the best way to structure an issue tree for data analysis planning?

Structuring an issue tree for data analysis involves breaking down a central question into mutually exclusive and collectively exhaustive branches, allowing you to enumerate specific hypotheses and map them to measurable KPIs.

How do I calculate sample size for an A/B test with specific KPIs?

Calculating sample size for an A/B test requires defining your KGI/KPI with quantitative success criteria, establishing the minimum detectable effect (MDE), and applying basic sample-size calculation guidance to ensure valid results.

Can I use CRISP-DM phase 0 problem-definition for actionable data decisions?

Yes, CRISP-DM Phase 0 problem-definition is designed for actionable data decisions by formalizing the objective, setting measurable KPIs, estimating required sample sizes, and producing next steps for data collection.

Does problem definition require documenting bias risks and execution logs?

Problem definition requires documenting bias risks and appending an execution log entry to maintain an analysis context file as the single source of truth, ensuring structured and transparent data-driven decision making.