Databricks adds zero-copy Workday Data Connect federation to Unity Catalog

How the connector works

The connector is a Lakehouse Federation feature that reads Workday's shared Iceberg tables directly from cloud storage. According to Databricks, queries run on Databricks compute, while Unity Catalog governs access with catalog-, schema- and table-level permissions, plus lineage and auditing for the foreign catalog.

Databricks describes the approach as catalog federation. Table metadata comes from Workday's Iceberg REST catalog, and Databricks compute then reads the data straight from Workday's managed object storage and runs the query itself. This differs from query federation, where SQL is sent to an external database and runs on that system's compute.

Access is read-only, so Workday remains the system of record for HR and finance data. Once the data is visible in Unity Catalog, teams can use Databricks SQL, AI/BI, notebooks and Genie alongside the rest of their enterprise data.

Setup and requirements

Databricks says enabling the connector is a short collaboration between a Workday administrator and a Databricks administrator.

• In Workday: enable Workday Data Connect for the tenant, share the tables to be queried through its Iceberg REST catalog, register an API client with a JWT Bearer Grant, upload the public key, and register an Integration System User with a role that grants read access to the shared tables.
• In Databricks: create one OAuth connection and a foreign catalog, then grant access and query. Databricks performs the OAuth token exchange with Workday's Iceberg REST catalog, so users do not manage a Workday token endpoint.

The requirements are a Unity Catalog enabled workspace and Databricks compute on Databricks Runtime 19 or above. Because the connector is in Beta, a workspace admin must enable it from the Previews page.

Three ways to bring Workday data into Databricks

The federation connector joins two existing options, and Databricks positions the three as complementary.

• Lakeflow Connect for Workday Reports, Workday HCM and Workday Activity Logs: fully managed, incremental ingestion into Delta tables in Unity Catalog. It suits durable copies, history tracking (SCD2) and scheduled pipelines.
• Workday Data Connect federation (new, Beta): zero-copy, read in place from data pre-ingested in Workday Data Cloud. It is aimed at customers using Workday Data Cloud who want to federate Workday's catalog into Unity Catalog.
• Live Data Query through a JDBC connection: on-demand queries fetched at query time, suited to real-time, ad-hoc lookups. It does not bring the underlying metadata into Unity Catalog.

The company notes that the new connector does not ingest or copy data into Databricks and is separate from the Lakeflow Connect ingestion connectors for Workday.

What teams can build, and why it matters

Databricks lists several use cases: asking natural-language questions about workforce and financial trends with Genie, combining Workday financial data with market, risk or sales data for forecasting and scenario planning, blending workforce data with operational metrics, and feeding governed, current Workday data into AI applications, models and agents.

Jonathan Rogers, Chief Information Officer at FactSet, said in the announcement that the connector "gives us a governed, zero-copy way to discover and catalog Workday data in Unity Catalog."

The release builds on the partnership announced at Workday Rising 2025, where Databricks was named a Workday Data Cloud launch partner. Databricks says it is one of the first of Workday Data Cloud's launch Data & AI platform partners to make a native Workday Data Connect federation connector publicly available. For data platform teams already using Workday Data Cloud, the change means HR and finance data can be queried under the same access controls as other Unity Catalog assets, without building and maintaining a separate pipeline.