Claude Data Analysis: From Clean Data to Clear Findings

Reusable prompt
Analyze this synthetic project table to compare elapsed calendar days by project type. First inspect data quality and propose cleaning rules. Show reproducible calculations, summarize group sizes, and suggest an appropriate chart. Separate observed differences from possible causes. Do not invent missing values or imply causation.

Review checklist
- Metric definitions and exclusions are explicit.
- Sample calculations agree with independent checks.
- Charts and conclusions describe the same dataset and units.

Suggested steps
1. Write the analytical question and define its metric. Identify the population, date range, units, and whether incomplete records belong in the calculation or a separate category.
2. Inspect the table for duplicates, missing values, inconsistent labels, and impossible dates. Document cleaning decisions so another reviewer can understand how the analysis dataset differs from the source.
3. Calculate a small sample independently before summarizing everything. Check duration calculations, denominators, and group definitions, then compare the verified sample with the proposed analysis method.
4. Choose a chart suited to the comparison. Show units and meaningful categories, avoid distorted axes, and explain when a small group or missing data makes a pattern uncertain.
5. Write findings separately from possible explanations. Include the checked numbers, their scope, and the questions you would investigate next before changing an operational process.

Source: AI News Pro — Fajad S