1. Home
  2. Spreadsheets & Data
  3. Mask Personal Data

Mask Personal Data

Mask names, phone numbers and ID numbers in columns, e.g. 010-****-5678.

For every format and option in one place, use Excel & CSV Workbench →

Runs in your browser · No upload
Press "Auto-detect column kinds" to fill the rules from headers and values. Masking cannot be undone, so keep the original file.

Operations

Mask names, phones, ID numbers, emails, addresses, cards and accounts, e.g. 010-****-5678.
Masking rules (pick a kind per column)
Masking cannot be undone. Keep the original file somewhere safe.
Files are processed in your browser and never sent to a server.

How to use

  1. Drop an .xlsx, .csv or .tsv file, or paste cells copied from Excel. This page opens with the Mask personal data step already added.
  2. Press Auto-detect column kinds to fill the rules from the header names (name, phone, email…) and the shape of the values. To do it by hand, use Add rule and pick a column and a kind.
  3. If needed, change the mask character per rule, or use Custom to set how many characters to keep at each end.
  4. Check the masked values (highlighted) and the step report (Masked 3 cells (이름 1, 전화 2) style) in the preview, then download.

Masking cannot be undone. Keep the original file somewhere safe. The tool never edits your original; it saves only a masked copy as a new file.

Default rules by kind

The default mask character is *. These results were checked by running the tool.

Kind Input Result Rule
Name 홍길 홍* 2 characters: keep the first
Name 홍길동 홍*동 3 characters: keep first and last
Name 남궁민수 남**수 4 or more: keep first and last, mask the whole middle
Name John Smith J*** S**** Latin names: keep only the first letter of each word
Phone 01012345678 010-****-5678 Normalized to hyphens, then the middle group is masked
Phone 02-123-4567 02-***-4567 As many marks as the middle digits
Resident ID 9001011234567 900101-1****** Birth date and the single gender digit only
Email hong@domain.com ho***@domain.com First 2 characters of the local part + ***, domain kept
Email ab@domain.com a***@domain.com If the local part has 2 characters or fewer, keep 1
Address 서울특별시 강남구 테헤란로 152 서울특별시 강남구 *** Keep region and district
Card 1234-5678-9012-3456 1234-****-****-3456 First 4 and last 4 digits
Account 110-123-456789 ***-***-**6789 Last 4 digits only, hyphens kept
Custom ABCDEFG (keep 2 and 1) AB****G Characters to keep as you set
  • Phone numbers are masked after being converted to the hyphen format, whatever the input looked like (digits only, spaces, +82). If a value cannot be read as a number, only its last 4 digits are kept and the other digits are masked.
  • Addresses are read with the same rules as Address Cleaner & Splitter. Region abbreviations stay as written (경기 수원시 영통구 광교로 107 becomes 경기 수원시 영통구 ***). A ZIP code is masked.
  • In Custom, if the characters to keep add up to the whole length or more, everything is masked, so nothing is exposed by accident.
  • Blank cells are left alone, and numeric cells such as 1012345678 are converted to text and masked.

How auto-detection decides

If a header contains one of these words, that kind wins (first match from the top). If the header says nothing, the kind is chosen when 80% or more of the first 200 values share one shape.

Kind Header words Value shape
Resident ID 주민, rrn, resident, ssn 900101-1234567
Email 메일, email a@b.com
Card 카드, card 16 digits
Account 계좌, account header only
Phone 전화, 휴대, 핸드폰, 연락처, phone, mobile, tel Korean phone number shape
Address 주소, address header only
Name 이름, 성명, 성함, 고객명, name header only

When two words overlap, as in 이메일 주소, the higher rule (email) wins. Auto-detection is only a suggestion, so look over the rules and add any column it missed. It will not find personal data in unlabeled columns or buried in a notes column.

Stacking with other steps

Order changes the result. For 서울 강남구 테헤란로 152, stacking split the address (add region and district) then mask the address leaves the address column as 서울 강남구 *** and the new columns as 서울특별시 and 강남구: the district is kept and only the detail is hidden. Masking first would leave only *** for the detail column after splitting, so keep the order split, then mask if you want to preserve the detail.

Stack Phone Number Formatter before masking to unify formats, which reduces values that are not recognized and so keep only their last 4 digits.

Things to watch

  • Even after masking, the remaining characters combined with other columns (birth date, region, employer) can identify a person. Keep only the columns your purpose needs and delete the rest.
  • The tool looks only at the shape of values. Personal data hidden in free-text columns such as notes will not be found.
  • The level of de-identification required by law differs by use. This tool only hides characters.

Your file is processed only in your browser and is never sent to a server.

Can't find the tool you need?

Tell us what you want to do. We build the most requested tools first.