Basics

What Counts as Personal Data?

Audience: founders, product, ops, engineering · Last reviewed: March 2026

See also: Compliance portal · Official resources · Guides index

Teams often underestimate how much data in their actual systems can relate to identifiable people. The mistake is usually not legal theory — it is workflow blindness.

Do not think only in terms of visible profile fields. Think about anything in your stack that relates to a real person and shapes records, decisions, communications, or service around them.

Where personal data shows up

  • Account creation and onboarding fields
  • Checkout, billing, or customer-service records
  • Support conversations and ticket histories
  • CRM and lifecycle marketing systems
  • Analytics and product events tied to accounts or identifiable users
  • Exports, spreadsheets, dashboards, and vendor tools built on that information

Why this matters operationally

Once businesses understand how broad their practical data footprint is, a lot of other pages on this site start to make sense: notices need to be more accurate, retention logic becomes more important, deletion becomes harder, and vendor review stops being optional theater.

Common mistakes

  • Ignoring support and CRM data because it is “not the product database”
  • Assuming analytics or event data is too abstract to matter
  • Forgetting spreadsheet exports and manual workflows
  • Failing to connect vendor tools back to the same underlying user/customer data

What to do next

The best next step is not to debate edge cases endlessly. It is to map the major systems and flows in the business so your team can see where personal data actually lives and moves.