Explore overview

Explore is where you analyse the tables your team has modelled, without writing SQL unless you want to. It’s built on Lightdash, so its own documentation covers the finer points of charts and dashboards.

Turning it on

An admin presses Enable Explore once per project. This checks the BigQuery connection and loads your dbt models. You need committed dbt code first (see Build). After that, Explore keeps itself up to date: every time a run of your committed code succeeds, new and changed models appear.

You don’t sign in again; Explore uses your platform login.

Working in Explore

1. Pick a table (one of your models).
2. Choose dimensions (things to group by, like country or date) and metrics (numbers, like users or revenue). They come from your dbt models’ YAML, so they mean the same thing for everyone.
3. Add filters, sort, and press Run query.
4. Switch to a chart, adjust it, and Save chart. Put saved charts on a dashboard, organised in spaces.

You can also create custom dimensions and metrics on the fly. They’re yours to explore with; to make one permanent and shared, add it to the model in Build.

Note

The usual Explore button for turning a custom dimension into a pull request is hidden here, because your code is managed in Build, not by Explore.

What your role allows

Account admins can build and manage content; editors can explore and build charts and dashboards; viewers can open dashboards and saved charts. The exact permissions follow your platform role, see Permissions.

The size of every query

Next to Run query you’ll see how much data a query would process, before you run it. See Query size and limits.