Best ChatGPT Prompt for Data Analysis
Turn messy data into clear insights — statistical analysis, chart recommendations, and actionable takeaways.
The Prompt
You are a senior data analyst with expertise in statistics, visualization, and business interpretation. Help me analyze some data. What I'm analyzing: [DESCRIBE THE DATASET] My sample data (10-30 rows) or description: [PASTE DATA OR DESCRIBE COLUMNS] My question: [WHAT DO YOU WANT TO KNOW — e.g. "what's driving churn?" or "are sales trending up or down?"] Tool I'll use: [Excel / Google Sheets / Python / R / SQL / BI tool] Audience: [executives / peer analysts / customers / personal use] Requirements: - Start with 3 key takeaways (before any details) - Suggest the most appropriate analysis type (comparison, trend, correlation, segmentation) - If statistical tests apply, name the test and why - Recommend 2-3 chart types to visualize the findings, with reasoning - Call out any data quality issues or missing context - Suggest 2-3 follow-up questions worth investigating - Translate findings into specific actions or decisions
How to Use This Prompt
- Paste real data when possible — make up sample rows if you're testing the prompt
- Be specific about the business question — 'what's going on' is too vague
- Tell the AI the audience — executives need different analysis than peer analysts
- Ask for the 'so what?' — every analysis should lead to a decision
Example Output
Dataset: Monthly churn rate and feature usage data for a SaaS product
Top 3 Takeaways:
- Churn is 3.2x higher among users who never used the integration features — integrations are a leading indicator of retention
- The top churn driver isn't price, it's time-to-first-value — users who don't complete setup in 7 days churn at 68%
- Customers who email support within 30 days have 40% lower churn — support contact correlates with retention, not frustration
Recommended analysis: Logistic regression on churn with usage features as predictors, plus a cohort analysis by onboarding completion.
Charts: Cohort retention heatmap (onboarding cohorts x months active), and a bar chart of churn rate by feature adoption tier.
Tips to Get Better Results
- Start with takeaways, not tables. Ask 'What are the 3 most important findings?' before the details.
- Correlation vs causation. Always ask 'Is this a real cause or just correlated?' before taking action.
- Compare to a baseline. Numbers mean nothing alone. Ask 'What's the benchmark for this metric in [industry]?'
- Sanity check. Ask 'What would make this analysis wrong?' — catches assumption errors.