Is removing names enough to anonymize business data?

Usually not. Combinations of details like exact dates, locations and rare attributes can still point to someone, so data is also generalized, aggregated and tested before release.

How do licenses protect against re-identification?

Licenses commonly forbid any attempt to re-identify people or companies and limit how the data may be used, on top of the cleanup done before delivery. This is general information, not legal advice.

Blog

What Is Re-Identification Risk? How Owners Keep Anonymized Data Anonymous

Q: What is re-identification risk?

The chance that anonymized records could be traced back to a real person or customer by combining the remaining details with other information.

By Andrew Smart, founder of MON$TER · About

Re-identification risk is the chance that someone could take records you believe are anonymous and work out who they describe, by matching them with other information. It is the main privacy question AI buyers and owners both care about before business data changes hands.

How it happens

Removing names is a start, not the finish. A record with an exact birth date, a rare job title and a small town might point to one person even without a name. The same goes for a customer account with an unusual size, location and start date. These combined details are called quasi-identifiers.

How businesses reduce the risk

Remove direct identifiers like names, emails, phone numbers and account IDs. See what counts as PII.
Generalize details: month instead of exact date, region instead of street address, size bands instead of exact counts
Aggregate small groups so no row describes one person or one customer. See what aggregated data is.
Drop rare outliers that stand out even after cleanup
Test before release: check whether any records are still unique on their remaining details
Contract limits: licenses commonly forbid any attempt to re-identify people or companies. See what if a buyer misuses licensed data?

Anonymized vs. pseudonymized

Swapping a name for a code is pseudonymization, and the risk stays higher because the link can be restored. Read anonymization vs. pseudonymization for the difference. Want to see personal details leave a file in real time? Try The Vault.

This is general information, not legal advice. Privacy rules differ by location and industry, so have your attorney confirm your approach.

Find out what your data is worth

Published MON$TER ranges: an exclusive one-time sale typically lands at $50K-$1M, and a non-exclusive recurring license typically runs $300K-$3M per year. No amount is guaranteed and not every company qualifies. Buyers see a description first, never your raw data. Get a free valuation at /estimate/ or start at Business Data.

← Back to the blog