Building dashboards
Turn a dataset into charts and collect them into a dashboard.
Publish a result to people who will not write a query. This page covers the Databaas specifics; for the full walkthrough, see the Superset dashboard guide.
Prerequisite
can_use_dashboards. See permissions reference.
The flow
Register a dataset
Point a dataset at a lakehouse table (a namespace and table reached through the database that connects to the lakehouse), or at a saved SQL query when you need a join or aggregation across tables.
Build charts
Create charts on the dataset: choose a chart type, pick the columns and metrics, and save.
Assemble the dashboard
Add your charts to a dashboard, arrange them, and publish it.
A chart runs its query as the viewer
Every chart re-runs its query each time someone opens it, under that person’s identity, not the author’s. Two people can open the same dashboard and see different numbers — each query returns only the rows that person is granted.
So a chart built on data a colleague has no grant for errors for them rather than showing an empty result. Building the dashboard does not share the data — the grant does. Make sure your audience has read access to the underlying tables; see data permissions.
Dashboards are a door, not a filter. Everyone who can open Superset sees the same datasets, charts and dashboards. What differs per person is which queries return rows. See how access works.