Gathid’s Identity Knowledge Graph

What Gathid’s graph models

Gathid models identity-relevant entities as nodes, including:

• People (employees, contractors)

• Non-human identities (service accounts, applications)

• Accounts

• Systems

• Groups

• Roles

These nodes are connected by explicit relationships such as:

• Membership

• Ownership

• Entitlement

• Linkage

The graph exists to reflect identity reality, not policy intent.

Relationship-first by design

Unlike relational systems, Gathid:

• Persists relationships directly

• Traverses them natively

• Reasons across multiple hops

This enables detection of:

• Indirect privilege

• Nested access paths

• Orphaned ownership

• Transitive risk

These signals cannot be reliably derived from tables alone.

Native data, not abstracted assumptions

Gathid ingests native identity data from source systems.

It does not rely on:

• Normalised abstractions

• Simplified role models

• Assumed hierarchies

If a system represents access in a messy or complex way, the graph reflects that reality — rather than hiding it.

Why rebuilds matter

Gathid rebuilds the identity graph from authoritative data on a regular cadence.

A rebuild:

• Clears the existing graph

• Reconstructs it from source truth

• Eliminates historical artefacts

This avoids:

• Drift

• Partial truth

• False confidence

The result is a current-state digital twin of identity.

Why traversal changes the questions you can ask

Instead of asking:

“Which accounts match these filters?”

Gathid asks:

“What access paths exist — and what do they imply?”

That shift is fundamental.

➡️ Next: What this enables in real partner engagements.