ChatGPT Prompts for Data Analysis

Write a SQL query for {{database_type}} to answer this question: "{{business_question}}"

Table schema:
{{table_schema}}

Table relationships:
{{relationships}}

Requirements:
- Use clear aliases for all tables and columns
- Add comments explaining each major section of the query
- Handle NULL values appropriately
- Optimize for readability over cleverness
- If the query requires aggregation, include a GROUP BY with the most useful breakdown
- Return results ordered by {{order_preference}}

After the query, explain what each section does and flag any assumptions you made about the data.

Write a {{language}} script to clean this dataset. Here is the structure:

Columns: {{column_definitions}}
Row count: approximately {{row_count}}
Known issues:
{{data_issues}}

For each issue, implement a cleaning step that:
1. Logs how many rows are affected before the fix
2. Applies the transformation
3. Validates the result

After cleaning, generate a summary report showing:
- Rows before/after
- Columns modified
- Values imputed or removed
- Any rows flagged for manual review

Use {{library}} for data manipulation. Add error handling so the script does not fail silently on unexpected values.

I need to visualize {{data_description}} for an audience of {{audience}}. The key insight I want to communicate is: "{{key_insight}}"

Data shape: {{num_rows}} rows, {{num_columns}} columns
Key variables: {{variables_list}}
Tool: {{visualization_tool}}

Recommend the top 3 chart types for this data and insight. For each recommendation:
1. Chart type and why it works for this specific data + insight combination
2. Which variables to map to which visual encodings (x-axis, y-axis, color, size)
3. Specific design choices (color palette for {{audience}}, annotations to highlight {{key_insight}})
4. Code snippet to create it in {{visualization_tool}}
5. One common mistake to avoid with this chart type

Rank them by how clearly they communicate {{key_insight}} to a non-technical audience.

I need to analyze whether {{hypothesis}} using the following data:

Dataset: {{dataset_description}}
Variables:
- Dependent variable: {{dependent_var}} ({{dv_type}})
- Independent variable(s): {{independent_vars}}
- Potential confounders: {{confounders}}
Sample size: {{sample_size}}
Significance level: {{alpha}}

Walk me through:
1. Which statistical test is appropriate and why (consider alternatives)
2. Assumptions to check before running the test, with code to check each one
3. The {{language}} code to run the analysis
4. How to interpret the output — what numbers matter and what they mean in plain English
5. How to report the results in a paper or presentation (APA format if applicable)
6. Limitations of this analysis and what could strengthen the conclusion

Design a dashboard for {{dashboard_purpose}} used by {{target_users}}.

Data sources available: {{data_sources}}
Key metrics to track: {{key_metrics}}
Update frequency: {{update_frequency}}
Tool: {{dashboard_tool}}

Provide:
1. Dashboard layout (which charts go where, with wireframe description)
2. For each chart/widget:
   - Metric shown
   - Chart type and why
   - Filters or drill-down capability needed
   - Thresholds or conditional formatting rules
3. Global filters (date range, {{filter_dimensions}})
4. A "north star" KPI that should be the most prominent element
5. Mobile considerations if applicable

Design for a {{expertise_level}} audience — choose complexity accordingly. The dashboard should answer these questions at a glance: {{key_questions}}

Write an executive summary and narrative for a data analysis report. Context:

Analysis topic: {{analysis_topic}}
Audience: {{audience}} ({{technical_level}} technical level)
Key findings:
{{key_findings}}

Business impact: {{business_impact}}
Recommended actions: {{recommendations}}

Structure:
1. Executive summary (3-4 sentences, lead with the most important finding)
2. Context paragraph (why this analysis was done)
3. Key findings section (use bullet points, each finding should have a data point and a "so what")
4. Implications and recommended actions
5. Methodology note (1-2 sentences, what data, what timeframe, what method)
6. Caveats and limitations

Tone: {{tone}}. Use specific numbers from the findings — no vague language like "significant improvement" without the actual percentage. Total length: 400-500 words.