Data cleaning in Excel, done for you with a change log
Exported data rarely arrives clean: names in mixed case, dates in three formats, numbers stored as text, stray spaces and half-empty rows. Attach the spreadsheet and this prompt has your todo.is agent standardize every column by your rules, flag what it can't fix, and hand back a clean copy with a sheet listing each change so nothing happens silently.
The prompt
- Clean the attached spreadsheet [FILE NAME]. Keep the original sheet untouched and add a "Clean" sheet. Rules: trim extra spaces, fix capitalization of names and cities, convert all dates to [DATE FORMAT], turn numbers stored as text into real numbers, standardize phone numbers to [PHONE FORMAT], split [COLUMN TO SPLIT] into separate columns, standardize values like "USA / U.S. / United States" to one spelling, and remove fully empty rows. Do not delete any row with data. Flag rows with invalid emails, impossible dates or missing required fields ([REQUIRED COLUMNS]) in a "Check these" sheet. Add a "Change log" sheet with counts per fix and examples. Return it as an .xlsx file.
What to change
- [FILE NAME]: Attach your Excel or CSV file and write its name, e.g. "event_signups.xlsx".
- [DATE FORMAT]: e.g. "YYYY-MM-DD" or "DD/MM/YYYY".
- [PHONE FORMAT]: e.g. "+44 7700 900123 (international)" or "leave phones as they are".
- [COLUMN TO SPLIT]: e.g. "Full name into First and Last" or "Address into Street, City, Postcode". Write "none" to skip.
- [REQUIRED COLUMNS]: e.g. "Email and Company".
Example result
- event_signups_clean.xlsx (example)
- Sheets
- • Original: your data, untouched
- • Clean: 2,418 rows, same order, cleaned
- • Check these: 57 rows that need a human decision
- • Change log: every type of fix with counts
- Change log
- • Extra spaces trimmed: 642 cells (e.g. " Maria Lopez " → "Maria Lopez")
- • Capitalization: 388 names and cities (e.g. "JOHN o'neill" → "John O'Neill", "new york" → "New York")
- • Dates unified to YYYY-MM-DD: 1,120 cells. The file mixed "03/04/2026", "4 Mar 2026" and Excel serial numbers. Dates like 03/04 were read as day/month because 92% of the clear dates in that column used day first
- • Numbers stored as text: 214 cells in "Tickets" and "Amount" converted
- • Phone numbers: 1,377 formatted to +44 style. 19 too short to fix, flagged
- • Country names: "UK", "U.K.", "England", "United Kingdom" → "United Kingdom" (233 cells)
- • Full name split into First and Last. 12 names with three or more parts kept the extra words in Last
- • Empty rows removed: 31 (no data in any column)
- Check these (sample)
- • Row 118: email "anna.berg@gmial.com", likely a typo of gmail.com
- • Row 402: signup date 2031-02-11, in the future
- • Row 977: Company missing (required)
- • Row 1,530: same email as row 1,529 with a different name, possible duplicate
- Not changed
- • No rows deleted except fully empty ones
- • Duplicates were flagged, not removed. Ask "remove duplicates by email" if you want that
How to do it with todo.is
- Attach your spreadsheet and copy the prompt with your formats and required columns.
- Send it on the Today screen in todo.is, or attach the file to your agent on WhatsApp or Telegram.
- Your agent cleans a copy, keeps the original sheet and logs every kind of change.
- Review the "Check these" sheet, then reply with decisions like "fix gmial to gmail everywhere".
Tips for a better result
- Always keep the original sheet. Cleaning is easier to trust when you can compare.
- Say which date order your data uses (day first or month first). 03/04 means different things in different countries.
- Ask for formulas instead of fixed values if you will paste new data in every month, e.g. TRIM, PROPER and TEXT.
- Make it recurring for weekly exports: "every Monday, clean the file I email you with the same rules".
data cleaning in Excel: FAQ
- What does data cleaning in Excel mean? Fixing inconsistent or wrong values so the data can be sorted, filtered and analyzed: spaces, cases, date and number formats, duplicates, blanks and typos.
- Which Excel functions help clean data? TRIM removes spaces, PROPER fixes name case, VALUE turns text into numbers, TEXT formats dates, and Find and Replace or Power Query handle bulk changes. Your agent can write these formulas for you too.
- Can it clean a CSV or Google Sheets file? Yes. Attach a CSV or export the Google Sheet, and ask for the result as .xlsx or .csv.
- Is my data safe? Your files stay in your own agent workspace and only your agent uses them.
JavaScript is required to use the todo.is app.