Canonical ID · Feature
Stop one person becoming several records across your team’s work.
A changed email, social profile or account can split notes, paths and follow-up across several records. Canonical ID brings supported evidence together and keeps uncertain pairs open for review instead of hiding the decision.
Canonical ID supports Network OS and Kernel. It is not a separate product.
Separate identifiers can split the same person’s notes, relationship paths and shared work.
Sam Rivera
3 linked identifiers
Do these belong to the same person?
The review keeps both source records visible before a person confirms or rejects the link.
New evidence or a later review can change the current resolution.
Evidence before certainty
Bring source records together without treating every hint as proof.
Signup, supported source sync, archive imports, enrichment, and model suggestions can contribute identity evidence. Candidate pairs stay separate from applied identity links, and uncertain matches remain open for a person to review.
Use supported evidence
Supported Graph One sources can contribute identifiers and matching evidence without making every hint final.
Keep uncertain pairs reviewable
A person can confirm, reject, skip, or later revisit a pair when Graph One needs help with the decision.
Use the resolved person in the work
Resolved people can support Search results, profiles, relationship paths, and person references in Tables or Boards.
Included in both products
Use the resolved person across Search, paths, and shared work.
Give the team one usable person behind the work
Network OS uses resolved people across Search, profiles, Tables, and Boards, with review when a pair stays uncertain.
KernelReview resolved people for use behind your own product
Kernel can review how current people, Search, Table, and Board services fit your product and what source work still needs scope.