Managing data interfaces

  • Release version: Australia
  • Updated March 26, 2026
  • 1 minute to read
  • Create a data interface to control how source data is filtered, combined, and exposed to consumers as a stable, governed schema.

    When you create a data interface, you select a combination method that determines how data from source tables is structured. The method you choose depends on where your data lives and what consumers want to query.

    If you select one source table, the combination method is automatically set to Single table. If you select two or more tables, you choose between JOIN and UNION in the combination method step. The method cannot be changed after the target table is created.

    Single table
    Exposes a subset of columns from one source table. Use this when all the data consumers need comes from a single table and you want to control which columns are visible.
    JOIN
    Combines related data from multiple tables that share a common key, using INNER JOIN. Use this when consumer data is split across tables and you want to correlate records by a shared identifier. Only records with matching values in all joined tables are included.
    UNION
    Stacks rows from multiple source tables with compatible schemas into a single view. Use this when the same type of data is distributed across systems or regions and consumers need a consolidated view. All rows from all tables are included.