Wall of Tools

Spreadsheet Anonymizer: Remove Personal Data from Excel and CSV

Replace names, emails, phone numbers, IDs and dates of birth in an Excel or CSV file with pseudonyms, hashes, masks or coarser values, column by column, before you share it.

  • Runs in your browser, nothing is uploaded
  • Free, no sign-up
  • 28 automated checks
Open or paste a spreadsheet to begin.

Remove personal details from a spreadsheet before you share it

A customer list sent to an agency, survey answers shared with a colleague, sales data for a report, a sample file for a software supplier, or a table pasted into an AI chatbot: often the people looking at it do not need to know who is who. Sharing less personal data is good practice everywhere and a legal duty under laws such as the GDPR. This tool goes through a spreadsheet column by column and replaces names, emails, phone numbers, IDs, addresses and dates of birth with stand-ins, so the data stays useful for counting, sorting and analysis without identifying anyone. The file never leaves your computer.

How to use it

  1. Open the spreadsheet or paste the cells, including the heading row.
  2. Check each column. The tool recognises common kinds of personal data from the headings and the values and suggests what to do; columns marked "personal" need attention. Change any suggestion with its drop-down.
  3. Look at the preview, which shows the first rows before and after.
  4. Download the result as Excel or CSV. Your original file is not changed.

The options

  • Pseudonym: each different value gets a stand-in such as Person 0007, used every time that value appears, so you can still count how often a customer ordered or join two columns. Emails become [email protected].
  • Hash: a code worked out from the value and a secret word. With the same secret word, the same value always gives the same code, even in another file next month, so datasets can be matched without revealing who is who.
  • Mask: keeps only the last few characters, such as an account number ending in 4821.
  • Year only, month only, first part only: keeps a coarser version, such as the year of birth instead of the date, or the first half of a postcode.
  • Round: amounts to the nearest 10, 100 or 1,000.
  • Shuffle: keeps every value but mixes them between rows, which preserves totals and averages while breaking the link to each person.
  • Remove: drops the column entirely. For free-text notes this is often safest.

Good to know

  • Pseudonymised is not always anonymous. A rare job title, a small town and an exact salary together can still point to one person. Remove or coarsen columns you do not need; the less you share, the safer it is.
  • The key file lists each original value with its stand-in, so you can match results back later. Anyone who has it can undo the pseudonyms: store it separately and never send it with the data.
  • Only the first sheet of an Excel workbook is used, and formulas are replaced by their values.

How this tool was tested

Before publishing, I ran 28 automated checks on the finished page in Microsoft Edge, which is built on Chromium like Chrome. Last full run: 15 November 2026.

  • Recognising columns from headings such as Full name, E-mail, Mobile, Date of birth, Customer ID, Postcode, Notes and Total spent, and from the values when a heading says nothing.
  • Every option: pseudonyms that stay the same for the same person whatever the capitals, example.com emails, masks keeping the last 4, years from ISO, day/month/year and Excel dates, first parts of UK, US and Canadian postcodes, rounding, shuffling (every value kept once), removing columns, and hiding emails and phone numbers inside kept text.
  • Hashes: the same value and secret word give the same code, also in another file, and a different secret word gives a different code.
  • Downloads: the Excel and CSV files were read back and searched for every original name, email and phone number, and none were found; the key file is a separate download.

What the tests did not cover: Only the first sheet of a workbook is used, and formulas become values. Whether a dataset is truly anonymous depends on what is left in it, as the page explains. Safari and Firefox were not tested. Something not working? Tell me on the contact page.

FAQ

Is my spreadsheet uploaded?

No. It is read and rewritten inside your browser.

Can I anonymise text instead, like an email or a chat?

Yes, with the AI privacy mask, which hides names and details in pasted text.

Will the same person get the same pseudonym in two files?

Pseudonyms are numbered per file. For codes that match across files, use Hash with the same secret word both times.

Published 16 November 2026