Skip to main content

Privacy matching workflow

Hash CSV Identifiers for Privacy-Preserving Matching

Hash normalized identifiers consistently while understanding dictionary attacks, salts and irreversible mismatches.

Use the matching tool

Hash Data

Open Hash Data

What the error actually means

Hashing can create a repeatable token for matching, but common values such as emails and phone numbers are guessable. Both parties must normalize identically, and stronger keyed hashing is preferable when an authorized shared secret is available.

Likely causes

  • Case and whitespace differ before hashing.
  • Phone numbers use different country-code formats.
  • Plain unsalted hashes are vulnerable to guessing.
  • The original identifier is still present elsewhere.

Consistent preprocessing

Problem

Hash Ada@Example.com and ada@example.com differently

Correct pattern

Apply an agreed normalization, then a governed hashing method

A safe repair workflow

  1. 1Agree on exact normalization rules.
  2. 2Choose the hashing or keyed-hash method.
  3. 3Test known pairs across both datasets.
  4. 4Remove unnecessary clear-text copies.

How to verify the result

A file that downloads successfully is not automatically a correct file. Check the result at both the structural and business-data levels:

  • Known identical IDs produce identical tokens.
  • Different IDs do not collapse unexpectedly.
  • The method and version are recorded.
  • Clear identifiers are handled securely.

Read the deeper guides

How to Anonymize CSV Data Without Ruining the DatasetA practical method for removing or replacing names, emails, IDs, dates, and other identifying fields while preserving useful CSV data.How to Clean CSV Data Without Hiding the Original ProblemsLearn practical techniques to clean messy CSV files. Remove duplicates, fix formatting, and standardize data without Excel crashes or complex scripts.