A good data analyst spends as much time writing SQL, cleaning data, and explaining results as analysing them. This pack gives you tested prompt templates for those tasks — query writing, debugging, Excel and pandas transforms, dashboard design, and turning numbers into a story your stakeholders act on.
How it works
Each prompt names the tool (SQL, Excel, Python), the input (your schema, columns, or
figures), and the output you want. The bracketed variables — [business question],
[table names], [audience] — give the model enough context to produce something you can run or
present. Filter by category or search, replace the brackets, copy, and paste into your AI tool.
Tips for better output
- Describe the schema, not the data. For SQL prompts, paste table and column names — the model rarely needs the actual rows and you avoid leaking sensitive records.
- Always test generated queries. Run them on a small sample first and verify the totals against a number you already know to be correct.
- Name the audience in storytelling prompts. “Explain this to a non-technical exec who cares about margin” produces a sharper summary than “summarise this.”
- Ask for the assumptions. Tell the model to state what it assumed about your data — it surfaces the exact spots where its query or interpretation could be wrong.