Querying Data Products
How querying a Data Product works, and where each query path leads.
Every activation path queries the same thing: the product's semantic layer. You write SQL or a structured query against semantic model names and their dimensions and measures; the engine resolves joins, aggregations, and access policies the same way described in Query in Studio. You never write raw warehouse SQL or define joins yourself.
This page is the thin map. Each query surface has its own page, and the full query reference lives in References.
How a query resolves
Your query -> semantic layer -> warehouse SQL (joins, aggregations, policies applied) -> resultTwo things follow from this:
You query logical model names and fields, not physical tables. Selected fields must be dimensions or measures the model exposes.
Measures are wrapped with
MEASURE()(or a matching aggregate). Dimensions are plain columns you select, filter, group, and order by.
Where to query
Validate fields and filters in the browser
Query in Studio (visual builder, Semantic SQL, GraphQL)
Query from a desktop SQL tool over the MySQL protocol
Call the product programmatically from code
Save a query for reuse and serving
For the full query reference, including the SQL form, aggregate functions, join syntax, filter operators, and the semantic query lifecycle, see the query reference in References.
Last updated
Was this helpful?