# Messy export to workbook prompt: free ChatGPT template > A copy-ready prompt that tells Claude exactly how to clean a messy CSV export into an Excel workbook with a data tab, a formula summary and a chart. Source: https://learn.ynteractive.com/content/prompts/messy-export-to-workbook-prompt · Updated 2026-10-14 · Free from Stepthrough (https://learn.ynteractive.com) [Reports and planning](https://learn.ynteractive.com/content/prompts#job-reports-and-planning) · [Finance](https://learn.ynteractive.com/content/prompts#industry-finance) · [Operations](https://learn.ynteractive.com/content/prompts#industry-operations) · Works in ChatGPT, Claude, Gemini and Copilot Updated October 14, 2026 · By [Robert Breen](https://learn.ynteractive.com/content/about) Use this when a system exports a file full of title rows, mixed date formats, dollar signs stored as text and repeated lines, and you need a clean workbook for a report. It says exactly what clean means, requires formulas and asks Claude to list every change so you can check its decisions. Upload the CSV first. It does not know what a duplicate means for your business, so read the list of removed rows yourself. Two identical rows can be a double export or two real sales. ## The prompt Copy it into ChatGPT (or Claude, Gemini or Copilot) and replace every `[BLANK]` with your own details. It uses the four parts from [Prompt Writing 101](https://learn.ynteractive.com/content/prompt-writing-101-role-context-task-format): Role, Context, Task and Format. Copy prompt ```text Role: You are a data analyst at [BUSINESS], preparing a clean workbook from the attached export. Context: The attached file is a messy export of [DATA TYPE]. Clean means one date format, amounts stored as numbers, no title rows, subtotal lines or repeated rows, and consistent category names. Use only the data in the file. Do not invent rows, and put any assumption you make where I will see it. Task: Build a Data tab with one clean row per [ROW MEANING], a Summary tab with [SUMMARY] using formulas and not typed-in results, and a chart of [CHART]. Then tell me what you changed. Format: An Excel workbook, followed by a short list of every fix, including each row you removed and why. Plain words, no jargon. ``` ## Fill in the blanks **`[BUSINESS]`** Your business name. **`[DATA TYPE]`** What the export contains, such as register sales or timesheets. **`[ROW MEANING]`** What one row should represent, such as one sale. **`[SUMMARY]`** The totals you need, such as total sales by month and category. **`[CHART]`** The comparison to chart, such as sales by month. ## Example, filled in **A made-up example** from the free lesson [Claude's Excel Skill: Turn a Messy Export into a Clean Workbook](https://learn.ynteractive.com/content/claude-excel-skill-messy-export-to-clean-workbook). Willow Creek Plant Shop is the made-up garden shop in the Claude's Excel Skill lesson, which starts from a messy sales export. The details below are an illustration, not the lesson's exact request. Copy example ```text Role: You are a data analyst at Willow Creek Plant Shop, preparing a clean workbook from the attached export. Context: The attached file is a messy export of register sales. Clean means one date format, amounts stored as numbers, no title rows, subtotal lines or repeated rows, and consistent category names. Use only the data in the file. Do not invent rows, and put any assumption you make where I will see it. Task: Build a Data tab with one clean row per sale, a Summary tab with total sales by month and category using formulas and not typed-in results, and a chart of sales by month. Then tell me what you changed. Format: An Excel workbook, followed by a short list of every fix, including each row you removed and why. Plain words, no jargon. ``` ## What a good answer looks like - Clean is defined in the prompt, so Claude does not have to guess your standards. - The summary uses formulas, so the totals update when the data changes. - Every removed row is listed with a reason, so you can restore a real sale. ## How to check it before you use it An AI draft can sound right and still say something you never told it. Use the habit from the [Reply Faster lesson](https://learn.ynteractive.com/content/chatgpt-reply-to-customer-emails): check every promise, name, date and number against what you gave it before anything goes out. - Click a few totals in Excel and read the formula in the formula bar. - Compare one month's total with the original export before you share the workbook. - Check that code execution and file creation is on in Claude's settings, or it cannot build the file. - Remove personal details the report does not need before you upload the file. ## Why this prompt works Cleaning data means making small decisions, so the prompt asks Claude to show them. Requiring formulas also makes the file auditable, since you can see how each total was built. ## Practice it in the free lesson **[Claude's Excel Skill: Turn a Messy Export into a Clean Workbook](https://learn.ynteractive.com/content/claude-excel-skill-messy-export-to-clean-workbook)**: Upload a messy sales export to Claude and get back a clean Excel workbook with a Data tab, a Summary tab built with formulas, and a chart, thanks to Claude's built-in Excel skill. You do every step yourself in a practice copy of Claude, and nothing touches your real accounts. [Start the free lesson](https://learn.ynteractive.com/modules/claude-skills-excel) ## Terms used here [CSV](https://learn.ynteractive.com/content/glossary/csv) · [Data extraction](https://learn.ynteractive.com/content/glossary/data-extraction) · [Code execution and file creation](https://learn.ynteractive.com/content/glossary/code-execution) ## Related prompts - [Data cleanup plan prompt](https://learn.ynteractive.com/content/prompts/data-cleanup-plan-prompt): Turn messy spreadsheet columns into clean, consistent ones: a column-by-column plan with standard values, rules for odd entries… - [Spreadsheet formula helper prompt](https://learn.ynteractive.com/content/prompts/spreadsheet-formula-helper-prompt): Get a Google Sheets or Excel formula for the job you describe, explained piece by piece, with a quick way to test it on a copy… - [KPI commentary prompt](https://learn.ynteractive.com/content/prompts/kpi-commentary-prompt): Explain a month's numbers in plain words: what moved, likely reasons you confirm, what to watch next, and no conclusions the… ## More reports and planning prompts - [Monthly cost share review brief prompt](https://learn.ynteractive.com/content/prompts/monthly-cost-share-review-brief-prompt): A copy-ready prompt that turns a client's monthly P&L totals into a one-page review brief: cost shares, what changed and three… - [NDA key terms summary prompt](https://learn.ynteractive.com/content/prompts/nda-key-terms-summary-prompt): Pull the key terms out of a nondisclosure agreement into one table with clause numbers, and flag anything beyond confidentiality… - [Plain-English document summary prompt](https://learn.ynteractive.com/content/prompts/plain-english-document-summary-prompt): Paste a contract, lease or policy and get a plain-English summary with section numbers, every date and dollar amount, and… ## Frequently asked questions **Can Claude open an Excel file as well as a CSV?** Yes, you can upload either. The Excel skill handles spreadsheet files, as long as code execution and file creation is switched on. **What if a total looks blank in the preview?** Download the file and open it in Excel. Older previews could leave formula results empty, though the file itself is fine. [All free prompt templates](https://learn.ynteractive.com/content/prompts) · [AI glossary](https://learn.ynteractive.com/content/glossary)