Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Glossary term

Personal Data

Any information relating to an identified or identifiable living person, such as a name, email address, location data or an online identifier.
StrategyconceptUpdated

Definition

The GDPR defines personal data as any information relating to an identified or identifiable natural person. A person is identifiable when they can be identified directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data or an online identifier, or to factors specific to their physical, economic, cultural or social identity. Two features of that wording do most of the work. "Relating to" covers information about what a person did, bought or said, not only who they are. "Indirectly" covers data that identifies nobody on its own but would do so when joined with something else the business, or someone else, holds.

Marketers often say PII, personally identifiable information. The phrase points in the right direction, but in everyday use it tends to mean only obvious identifiers such as names and email addresses. The regulation's term is broader and is the one that decides which rules apply.

In practice: where it turns up in marketing systems

Personal data enters marketing through planned routes and accidental ones. The planned routes are familiar: the CRM, the email platform, customer lists uploaded to ad platforms, sign-up and lead forms, loyalty records. The accidental routes are the ones that cause trouble, because nobody decided to collect anything. The guide to keeping personal data out of analytics lists them: form fields carried into page addresses, names in page titles, email addresses typed into site search, campaign links filled with a subscriber's details by an email tool.

The same pattern is now repeating with AI tools. The guide to AI disclosure and data privacy sorts material into public or non-personal data, personal data and special-category data, and allows personal data only into approved tools with terms that forbid training on inputs. The routine for minimising data before prompting then reduces an export or transcript set to what the task needs before it is pasted anywhere.

Personal data also appears in places that are not systems at all. A social media policy has to tell staff to keep customers' and colleagues' names, faces and stories out of posts, and in B2B sales the late-stage deal content a buyer asks for includes how the vendor will handle personal data about their staff or customers, an answer the privacy owner writes rather than marketing.

Identified, identifiable, pseudonymous and anonymous

These four states describe how close data is to a named person.

  • Identified: the data names the person or reaches them directly, such as an email address or phone number.
  • Identifiable: the data does not name anyone, but the person could be singled out with other information, such as an internal customer number, a cookie identifier or a precise location history.
  • Pseudonymous: identifiers have been replaced with codes, and the key that links codes to people is kept separately. The GDPR treats pseudonymised data as personal data still, because the link can be restored.
  • Anonymous: the information can no longer be related to an identifiable person by any reasonably likely means, such as a properly aggregated count. Only this state falls outside the definition.
Illustrative example — not a real result

A dental group finds patients' email addresses in its page URLs

Drummelow Dental Group runs nine practices and takes bookings through a web form. Reviewing landing page reports, a marketing assistant notices that the confirmation page appears as hundreds of separate rows, each with a different address. The form used the GET method, so every submitted field travelled in the URL: 640 confirmation page views over five weeks carried a patient's email address and phone number into analytics. The team changes the form to submit its fields in the request body, adds a rule that strips the email and phone parameters from page locations before they are sent, and asks the practice manager who owns privacy to record the incident and decide on clean-up. A check a fortnight later finds no new addresses.

The GDPR is the regulation whose definition this page follows, and it sets the rules that apply once data counts as personal. Every use of personal data needs a Lawful Basis chosen for its purpose. Most personal data a marketing team handles is First-Party Data, collected directly from its own customers, which makes it valuable and makes the team responsible for it. Hashing encodes identifiers such as email addresses for matching on ad platforms; it changes the format of the data, not its status.

Common confusions

  • Personal data means names and email addresses. Online identifiers, location data, customer numbers and behavioural records about an identifiable person count too.
  • Hashed or pseudonymised data is anonymous. Anyone holding the original value or the key can link it back, so it remains personal data.
  • Business contacts are not personal data. A named employee's work email address still relates to an identifiable person. Separate rules on unsolicited electronic marketing treat individuals and corporate subscribers differently, which the ICO's direct marketing guidance covers, but that difference does not take the contact outside data protection.
  • Consent turns any collection into a safe one. Consent is one possible lawful basis; it does not justify collecting more than the purpose needs or storing personal data in a reporting tool whose terms forbid it.
  • Data spread across systems is not personal. A record that identifies nobody alone can identify someone once it is joined to the CRM, which is exactly what marketing systems are built to do.

More terms to look up

Where this term is used

Sources

  1. Regulation (EU) 2016/679 (GDPR) — EUR-Lex
  2. Direct marketing guidance — Information Commissioner's Office (UK)

Related