Discovery
Scans databases, tables, and columns using read-only `SELECT` access.
Ketch and Databricks
Find where personal data lives inside a lakehouse platform, and fulfill deletion requests against it directly, managed through the same role-based access control already governing everything else in the workspace.
About Databricks
Databricks runs analytics and machine learning workloads across data lakes and warehouses at once, holding data that can span both raw and heavily transformed states. Managing access to that data through the platform's own RBAC model, rather than a workaround, keeps a privacy integration consistent with how every other access decision in the workspace already gets made.
Ketch connects to Databricks through a Ketch Transponder with that consistency in mind.
Capabilities
How Ketch works with Databricks
The Ketch Databricks integration covers Discovery and Rights Orchestration for both Right to Access and Right to Delete.
Rights Orchestration
Requires `UPDATE` and `DELETE` access granted on top of that, scoped to specific tables and columns. Access is controlled through Databricks's own role-based access control, governing permissions on SQL warehouses, catalogs, and tables the same way every other access decision in the workspace is made.
With Ketch, teams can
- Discover which Databricks tables hold personal data using read-only access
- Grant rights-execution privileges separately from Discovery, deciding independently whether and when to authorize direct deletion and update
- Manage the Ketch service account through Databricks's own RBAC model, consistent with how every other access decision in the workspace is made
The gap
The problem this integration solves
A lakehouse platform spanning both raw and transformed data still needs personal data actually located and reachable through the platform's own access model:
01. Manually reviewing tables and columns for personal data across a lakehouse doesn't scale and goes stale as data pipelines evolve
02. A privacy integration managed outside the platform's own RBAC model creates an inconsistent access pattern worth avoiding
03. Fulfilling a deletion request against a lakehouse has historically meant a manually written query, slow and hard to prove complete
Ketch resolves this by connecting through Databricks's own RBAC and access token model, and by separating discovery from execution as two distinct privilege grants.
Why Ketch
Why teams choose Ketch for Databricks privacy compliance
Permissioning infrastructure that governs Databricks the same way it governs every other system in your stack — not a one-off connector bolted onto a banner.
Managed through Databricks's own RBAC
The Ketch service account's permissions on SQL warehouses, catalogs, and tables are governed the same way every other access decision in the workspace is.
Least-privilege by design
Discovery requires only `SELECT`; rights execution is a separate, explicit grant.
Backed by enforcement precedent
Regulators increasingly expect businesses to prove technical enforcement, not just describe it on paper, the same underlying expectation that applies to knowing what personal data a lakehouse holds.
Questions about the Databricks integration
Related reading
Expert insights for teams connecting Databricks
Integrations
Pre-built APIs with 1,000+ systems, apps, and models
Ketch ships connectors and SDKs so consent, rights, and policy flow into your CDPs, warehouses, ad platforms, and AI stack — without a custom data pipeline.
See Databricks permissioning running end to end
Book a demo to walk through rights, consent, and preference orchestration on your stack — or start free and connect Databricks yourself.
Get Started Free
Get started in less than 5 min