Data & analyticsData qualitytext

Find the story and errors in a messy dataset

Audits a CSV for quality issues before building defensible findings and charts.

Ready to use

Prompt

Analyse the attached dataset for [DECISION OR QUESTION]. First describe the schema, row count, units, missing fields, date range, likely keys and any sensitive columns. Do not infer what a column means from its name alone; ask or label assumptions.

Profile duplicates, impossible values, inconsistent categories, outliers and selection bias. Show a cleaning plan with before/after counts and keep a reproducible transformation log. Only then calculate the metrics needed for the question. Separate descriptive findings, plausible explanations and claims the data cannot support. Propose three charts with exact axes, aggregation and caveats; do not use a chart that hides uncertainty.

Return an executive summary, data-quality table, calculation notes, chart specifications and five questions for the data owner. Do not fabricate missing rows or treat correlation as causation.