Data Anonymization

Data anonymization removes or alters personal details in a dataset so the people in it can't be identified, letting the data be used for analysis, testing or sharing with less privacy risk.

IntermediateCategory: Data protection, encryption and privacy

Also called: anonymization, de-identification, pseudonymization, data masking

What is data anonymization?

Anonymization transforms data so it no longer identifies anyone. A customer database might become a dataset of purchase patterns by age band and region, with names, emails and exact addresses removed. NIST calls the broader process de-identification and warns that it's harder than it looks.

Common techniques

  • Removing direct identifiers: names, emails, phone numbers, account numbers.
  • Generalizing: exact birth dates become years or age ranges; addresses become ZIP code prefixes.
  • Masking: showing only part of a value, such as the last four digits.
  • Adding noise or aggregating: reporting totals or slightly altered values instead of individual records.
  • Pseudonymization: replacing identifiers with codes. This is reversible if the code key exists, so laws like GDPR still treat pseudonymized data as personal data.

Why true anonymization is hard

NIST's report explains that people can often be re-identified by combining "anonymous" data with other information. A ZIP code, birth date and gender together can single out many people. For health data, HHS sets out two HIPAA methods: Safe Harbor, which removes 18 specific identifiers, and Expert Determination, where a qualified expert confirms the re-identification risk is very small.

Uses for businesses

  • Testing and development with realistic but non-identifying data.
  • Analytics and reporting.
  • Sharing data with partners or researchers.

Anonymize before data leaves its secure home, keep only the fields you need, and treat pseudonymized data as still sensitive.

Related terms

Sources

  1. NISTIR 8053: De-Identification of Personal Information (NIST)
  2. Guidance Regarding Methods for De-identification of PHI (U.S. Department of Health and Human Services)

Drafted with AI assistance, then edited and fact-checked by a person. How we use AI.