- Direct signals (what you actively do)
- Search queries & site visits: If you search for “running shoes” on a search engine or visit product pages on retailers, those queries and pageviews are recorded by the search engine, publisher, or advertiser.
- Form fills, logins, and emails: If you give an email/phone to a site (newsletter, account, order), that’s strong deterministic data.
- Tracking technologies
- Cookies & tracking pixels: Small files or 1×1 images on web pages that tell ad networks which pages you visit. Common on e-commerce and publisher sites.
- Mobile SDKs & app IDs: Apps include software-development-kits that send events (searches, product views) linked to mobile advertising identifiers (IDFA on iOS, GAID on Android).
- Browser / device fingerprinting: Collects device/browser attributes (screen, fonts, time zone) to create a unique fingerprint when cookies aren’t available.
- URL parameters & referer data: Product identifiers or search terms sometimes appear in URLs and are captured by analytics/ad scripts.
- Cross-site / cross-app linking
- Cookie-syncing & ID graphs: Ad partners exchange identifiers (or sync hashed emails) to stitch a single identifier across sites/apps so the same user can be recognized elsewhere.
- Deterministic matching: Exact matches (e.g., hashed email) across datasets give high confidence.
Probabilistic matching: Statistical signals (behavior patterns, IP + device) used to guess identity when exact match is absent.