Where is the output saved?

The skill updates or creates an 'analysis_context.md' file, which serves as the 'source of truth' for the entire data analysis lifecycle.

Does this skill require existing data to work?

No, data-define is meant to be used before data processing. It helps you define what data you need to collect and how you will eventually measure it.

Can it help with statistical planning?

Yes, it includes built-in logic to estimate necessary sample sizes using power analysis, ensuring your analysis has enough data to be statistically significant.

When should I use the data-define skill?

Use it at the very beginning of an analysis project (Phase 0) when you need to clarify goals, structure a problem, or decide which metrics will determine success.

How does it improve KGI/KPI design?

It provides a standardized framework including definition formulas, data sources, and quantitative success criteria like Minimum Detectable Effect (MDE).

Data Analysis Problem Definition

Name: Data Analysis Problem Definition
Author: mutsumi-yamamoto

bymutsumi-yamamoto

0•

데이터 과학 및 ML

Defines business problems, designs KGI/KPI frameworks, and establishes testable hypotheses for data-driven decision making.

The data-define skill automates the critical initial phase of data analysis by transforming vague business objectives into structured, actionable problem statements. Using an issue-driven approach, it builds MECE (Mutually Exclusive, Collectively Exhaustive) issue trees, defines measurable KGI/KPIs with statistical rigor, and outlines specific hypotheses for validation. It ensures all analysis is actionable by identifying potential biases upfront and calculating required sample sizes, preventing wasted effort on non-statistical or non-impactful data exploration.

주요 기능

01Hypothesis planning with validation methods and falsification criteria

02Comprehensive bias risk assessment (selection, survivorship, and confirmation bias)

030 GitHub stars

04Statistical power analysis for sample size estimation using statsmodels

05Strategic KGI/KPI design including definitions, formulas, and success thresholds

06Issue-driven problem structuring with MECE issue tree construction

사용 사례

01Converting high-level business goals like 'increase revenue' into specific, testable analytical questions

02Designing a formal measurement framework for new product features or marketing campaigns

03Structuring complex organizational problems before starting exploratory data analysis (EDA)

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mutsumi-yamamoto/claude-data-analysis-marketplace data-define

For use in Claude.ai and ChatGPT

주요 기능

01Hypothesis planning with validation methods and falsification criteria

02Comprehensive bias risk assessment (selection, survivorship, and confirmation bias)

030 GitHub stars

04Statistical power analysis for sample size estimation using statsmodels

05Strategic KGI/KPI design including definitions, formulas, and success thresholds

06Issue-driven problem structuring with MECE issue tree construction

사용 사례

01Converting high-level business goals like 'increase revenue' into specific, testable analytical questions

02Designing a formal measurement framework for new product features or marketing campaigns

03Structuring complex organizational problems before starting exploratory data analysis (EDA)

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mutsumi-yamamoto/claude-data-analysis-marketplace data-define

For use in Claude.ai and ChatGPT