Databricks Describes a Workday Connector That Queries Data Without Copying It
The connector runs queries on Databricks compute while Workday remains the system of record. The announcement’s date leaves current beta availability unconfirmed.
Listen to this story
The audio brief
Story brief
3 key pointsThe connector gives Databricks users read-only access to Workday tables for analysis while leaving the data in Workday-managed storage. Unity Catalog applies permissions, lineage and auditing, so teams can combine governed HR and finance data with other enterprise information without first ingesting it. It is aimed at live analysis rather than preserving historical snapshots; Databricks recommends Lakeflow Connect for durable copies. The announcement labels the connector beta, but its October 8, 2026 date is later than the page’s October 7 capture, so availability is unconfirmed.
- 01
Databricks federates table metadata from Workday’s Iceberg REST catalog, then runs queries on Databricks compute against the source tables.
- 02
The read-only connection cannot write changes to Workday, which remains the system of record.
- 03
Beta setup requires a Unity Catalog-enabled workspace, Databricks Runtime 19 or later, and configuration by both companies’ administrators.
Databricks describes a new Workday Data Connect federation connector that lets teams analyze shared HR and finance tables without copying them into Databricks. The connector brings Workday data under Unity Catalog’s access controls and makes it available alongside other enterprise data for analytics and AI workflows, according to the company.
The announcement calls the connector publicly available in beta, but is dated October 8, 2026—after the page was captured on October 7. That mismatch leaves current availability unconfirmed.
Databricks runs the query; Workday keeps the data
The connector uses catalog federation: it retrieves table metadata from Workday’s Iceberg REST catalog, the interface describing its shared tables. Databricks compute then reads those tables directly from Workday’s managed cloud storage and executes the query. That differs from query federation, which sends SQL to an outside database to run there.
Access is read-only, so the connector does not write changes back to Workday. Workday remains the system of record for HR and finance data. Unity Catalog supplies permissions at the catalog, schema and table levels, along with lineage and auditing to track data use.
Workforce questions and financial planning
Once administrators grant access, teams can use Databricks SQL, notebooks, AI/BI and Genie to explore the federated data. Genie supports natural-language questions about workforce and financial trends. Databricks gives examples such as comparing headcount growth across cost centers or examining attrition by region and job family.
The company also describes joining Workday financial data with market, risk or sales information for forecasting and scenario planning. Workforce data can be combined with operational metrics, while governed Workday data can feed downstream AI applications and agents. These are proposed uses, not measured improvements in forecasting or AI accuracy.
Two administrators, one connection
Setup requires coordination between Workday and Databricks administrators. On the Workday side, an administrator enables Data Connect, shares approved tables and configures an API client and an Integration System User with read access. The Databricks administrator creates an OAuth connection and a foreign catalog—a catalog representing externally stored tables—then grants user access. Databricks handles the OAuth token exchange with Workday.
Listed beta requirements
- A Unity Catalog-enabled workspace.
- Databricks compute running Databricks Runtime 19 or above.
- A workspace administrator must enable the connector from the Previews page.
Sources
- databricks.comAnnouncing Workday Data Connect federation in Unity Catalog
Reader comments
Newest comments first. Replies stay oldest first.