Anonymisation
The irreversible process of altering data so that individuals can no longer be identified, directly or indirectly.
Full Definition
Anonymisation is the process of permanently modifying personal data so that it can no longer be attributed to a specific individual, even with additional information. Unlike pseudonymisation, true anonymisation is irreversible. Once data is genuinely anonymised, it falls outside the scope of most data protection laws including GDPR. Achieving true anonymisation is technically challenging — many supposedly anonymised datasets can be re-identified using auxiliary information. Common anonymisation techniques include data aggregation, data suppression, noise addition, and generalisation.
How each law treats it
Anonymised data is excluded from the GDPR's scope because the data subject is no longer identifiable. The regulator reads the test strictly — pseudonymised data is still personal data.
The DPDP Act applies only to digital personal data that can identify an individual. Properly anonymised data falls outside the Act, but the Data Protection Board treats re-identification risk as an operational burden on the Fiduciary.
Defines 'deidentified' with four engineering tests (technical safeguards, administrative controls, contractual prohibitions, no attempt to re-identify). Meeting all four removes the data from CCPA scope.
In practice
- K-anonymity (ensuring every row shares its quasi-identifiers with at least k-1 others) and differential privacy are the two most widely accepted techniques — both are harder than they look.
- Aggregation to the group level (counts, averages, histograms) is usually safe when the smallest group is large enough; the ICO publishes a working floor of ~10 individuals.
- Removing names and emails is not anonymisation — a quasi-identifier set (date of birth + ZIP + gender) re-identifies 87% of the US population per Sweeney's 2000 study.
Common pitfalls
- Calling pseudonymised data 'anonymous' and relying on the label to escape regulatory scope.
- Shipping anonymised exports to third parties without a contractual re-identification prohibition — the CCPA engineering test explicitly fails if the contract is missing.
- Treating anonymisation as a one-shot — if the surrounding environment changes (a new linkable dataset is published), yesterday's anonymised data may be re-identifiable today.
Frequently asked questions
Is hashing enough to anonymise an email address?+
No. A SHA-256 of an email address is pseudonymisation — anyone who knows the plaintext can reproduce the hash and link the dataset. Add salting and keep the salt secret, or use a bucketed cohort id instead.
Does anonymisation remove data entirely from GDPR scope?+
If the anonymisation is irreversible, yes. Recital 26 says data rendered anonymous 'in such a manner that the data subject is not or no longer identifiable' is outside the GDPR. The question is whether the anonymisation is actually irreversible given the data environment.
What's the difference between anonymisation and deidentification under the CCPA?+
Terminology. CCPA uses 'deidentified' and sets a four-part engineering test; the result is comparable to GDPR anonymisation but the specific technical obligations are different. Design to the stricter of the two if you operate across both.
Related terms
Pseudonymisation
The process of replacing directly identifying data with pseudonyms, so the data can no longer be attributed to a specific individual without additional information.
Personal Data
Any information that relates to an identified or identifiable natural person.
Data Minimisation
The principle that only personal data that is adequate, relevant, and limited to what is necessary should be collected and processed.
Relevant regulations
Automate your privacy program
TruePrivacy handles DSRs, consent management, data mapping, and breach response — all in one platform.
Free 14-day trial · No credit card required · Setup in minutes