Topics define the business areas Zoë and Explore can analyze. A topic groups related views together, such as orders, customers, products, or subscriptions, so users can ask questions without needing to understand the underlying warehouse tables.

Use a topic when a set of tables belongs together and should be explored as one business area.

A topic tells Zenlytic:

Good topics make questions easier to answer because they narrow the context to the right part of the data model.

Common topics include:

A topic should map to how users think about the business, not just how tables are stored in the warehouse.

Keep topics focused. One clear business area is better than one large topic that includes everything.

Use clear labels and descriptions. The topic name and description should help users understand when to use it.

Include the fields users actually need. Avoid exposing every technical column if it is not useful for analysis.

Make joins explicit. If a topic spans multiple views, confirm the joins and identifiers are correct before relying on it for analysis.

Test with real questions. After creating or updating a topic, ask Zoë common customer questions and confirm the answers use the expected fields and joins.

Update a topic when:

Topics are part of the semantic layer. If you need exact syntax or YAML reference details, use the Zenlytic docs. The Help Center should explain what topics are and when to use them; the docs should hold the implementation details.