Skip to main content

​
What it is

The Contacts dataset is one row per contact created in your space within the window. A contact is a person known to your tracker (internally called a profile).
  • Grain: one row per contact (id).
  • Windowed on: created_at (UTC timestamp) — the contact’s original creation time. The internal row timestamp is never exported.
  • Path segment: contacts.
Contacts are mutable. Each run carries a contact’s state as of run time, so re-runs self-heal updates for contacts created inside the window. A contact created before the window is not re-exported when it later changes — for a full refresh, run a one-time export over all history.
Contacts flagged internally as pathological (runaway identity graphs) are excluded outright.

​
Columns

​
Contact

​
Identity

​
Location & device

​
Timestamps

​
Company

​
Custom fields

Your space’s custom contact fields, by slot. The slot number determines its type: All optional. The create form shows only the slots your space has configured, labeled with each field’s name.
Output naming is your choice at creation. Custom-field columns can be named by their stable slot (profile_field_7) or by the field’s label (frozen as of creation). With label naming, renaming a field later does not change an existing export’s columns.

​
Identity graph

The contact’s full multi-value identity, exported as JSON arrays. All optional and default to off. Values inherited from contacts merged into this one are included, and the lists are uncapped (unlike the Public API’s recent-5 token list).