Dashboards are built from widgets, each backed by a metric or set of metrics from your connected data sources. This article covers the basics of building one from scratch.
Starting from a template versus starting blank
If a prebuilt template exists for your data sources and use case (for example, a blended GA4 and Meta Ads template), starting from it and customizing is usually faster than building from a blank dashboard. Templates are a starting point, not a fixed layout — every widget can be edited, removed, or added to after the template is applied.
Adding a widget
Each widget is configured with a data source, a metric, a visualization type (like a line chart, a single-number stat, or a table), and an optional filter or segment. Widgets from different connected sources can live side by side on the same dashboard, and some widget types support blending a metric across more than one source directly.
Organizing a dashboard
A dashboard that tries to show everything tends to become one nobody reads closely. A few practical guidelines:
- Lead with the two or three metrics that matter most for this dashboard's specific purpose, placed at the top.
- Group related widgets together rather than scattering related metrics across the layout.
- Avoid duplicating the same metric across multiple widgets on the same dashboard unless you're intentionally comparing two segments side by side.
Sharing a dashboard
Once built, a dashboard can be shared with other members of the workspace based on their role, or shared externally — see Sharing and Embedding Dashboards for details on external sharing and embedding options.