ChatGPT / 4 min
ChatGPT spreadsheet prompts: analyze Excel, CSV, and Google Sheets with clearer instructions
A practical ChatGPT spreadsheet prompt guide for formulas, data cleaning, summaries, and chart planning in Excel, CSV, and Google Sheets. Define columns, outputs, and verification rules first.
Watch on YouTube
BananaNL in 1 minute
A quick look at how selected prompts move into NotebookLM and AI Chat input fields.
Watch on YouTubeDo not upload the table before you define the question
ChatGPT can help with Excel, CSV, and Google Sheets work: summarizing data, writing formulas, cleaning messy rows, and planning charts. But a prompt like “analyze this spreadsheet” leaves too much room for guessing about the columns, time period, and business decision you care about.
OpenAI's Data analysis with ChatGPT guidance recommends structured data, clear column headers, one record per row, and explicit instructions about the columns, calculations, groupings, or chart types you want. A good spreadsheet prompt is therefore a short analysis brief, not only a file attachment.
When a spreadsheet task repeats, save the prompt pattern in BananaNL so you can reuse it for sales tables, survey CSVs, budget trackers, and reporting sheets with only the column names changed.
Five decisions to include in a spreadsheet prompt
- State the goal in one sentence: formula help, outlier detection, data cleanup, meeting summary, or chart planning.
- Explain the columns: names, units, date format, blank values, IDs, and category labels.
- Limit the analysis scope: target sheet, period, segment, category, or grouping level.
- Lock the output format: Excel formula, table, bullet summary, chart recommendation, or follow-up questions.
- Ask for verification details: assumptions, skipped rows, formulas used, and anything that was inferred rather than observed.
The four inputs to prepare first
Where this prompt pattern helps
| Formula writing | Generate or fix SUMIFS, XLOOKUP, FILTER, percentage-change, and lookup formulas from column names and conditions. |
|---|---|
| Data cleaning | Find duplicates, blanks, inconsistent dates, mixed units, and category naming issues before analysis begins. |
| Analysis | Summarize sales, ads, surveys, inventory, or budgets by month, category, region, owner, or campaign. |
| Reporting | Turn the findings into a meeting-ready summary, chart plan, and next questions instead of a raw list of numbers. |
Spreadsheet prompts to try in ChatGPT
Create an Excel formula from column names
For the attached table, the goal is to calculate monthly sales totals by category. The relevant columns are Date, Category, and Sales Amount. Write an Excel formula using placeholder ranges, then explain the formula logic and assumptions in three short bullets.
This keeps the answer tied to the named columns and gives you enough explanation to verify the formula before using it.
Prepare a CSV cleanup checklist
Before analyzing this CSV, check for duplicate rows, blanks, date-format issues, amount-unit issues, and inconsistent category labels. Return a table with the likely issue, how to check it, and the safest fix. If the data does not prove something, mark it as needs verification.
It makes data quality visible before you rely on trends or totals that may change after cleaning.
Create a meeting summary and chart plan
Review the attached sales table for a weekly business meeting. Summarize the overall trend, categories that are growing, categories that are declining, and follow-up questions in no more than three bullets each. Then recommend the best chart type and explain why.
The prompt moves from analysis to a usable reporting format, which is usually what the spreadsheet work was for.
Save spreadsheet prompt patterns in BananaNL by use case
Spreadsheet prompts differ by use case: formulas, cleanup, dashboards, reporting, and chart planning each need slightly different checks. Saving the patterns lets you swap only the column names and output goal instead of rewriting everything.
BananaNL is a Chrome extension that inserts saved prompts into AI Chat tools such as ChatGPT, Gemini, and Grok. It does not auto-submit, so you can attach the spreadsheet, review the column names, and adjust the prompt before sending. NotebookLM usage starts free, while AI Chat integration is a paid feature.
FAQ
What should I write first when using ChatGPT with Excel?
Start with the goal, target columns, and output format. Column names and units are especially important because they control how ChatGPT interprets the numbers.
Can I upload a large spreadsheet as-is?
Large or messy files may not be analyzed completely. Narrow the sheet, time period, and columns, or split the task into smaller checks when accuracy matters.
Can I use ChatGPT's analysis without checking it?
For important decisions, review the formula, assumptions, excluded rows, and source row counts. If ChatGPT used code-based analysis, ask it to show the method and outputs before relying on the result.
If searching for prompts is the hard part, use BananaNL
Prompts become useful when they are close to the input field. Use BananaNL to carry them there, then adjust before sending.