Skip to main content
Version: V3.2

Create & Edit Dashboards

Capture dashboards are built in Grafana. In practice, you will spend most of your time doing three things:

  • creating a dashboard
  • adding or editing panels (the visualizations)
  • refining queries and variables (the “logic” behind the visuals)
Grafana basics

The editor you see is Grafana’s dashboard editor.
Capture provides access control, company scoping, and (optionally) Fleet/Project sharing on top of it.

Version note

Button labels and layouts may vary slightly by Grafana version.
The workflow stays the same: create → add panel → write query → apply → save.

Create a dashboard​

Start by creating/searching a folder where the dashboard should be placed (middle column). You can then add a dashboard (+ in the top right corner) in the selected folder.

  1. Open the dashboard you created
  2. Click Add a visualization
  3. Choose a data source
  4. Build a query (Grafana query editor)
  5. Click Save (top-right)
Save behavior

Panel changes are not persisted until you Save the dashboard.

Add a panel (to an existing dashboard)​

Panels are the building blocks of dashboards: charts, stat tiles, tables, etc.

  1. Open an existing dashboard
  2. Click Add visualisation (typically in the top right when you are in 'edit' mode)
  3. Select a data source & write your query
  4. Save the dashboard

Dashboards Create

Edit a panel​

  1. Hover the panel and click the three dots
  2. From the dropdown menu, select Edit
  3. Adjust:
    • query
    • visualization type
    • field options (units, decimals, thresholds)
    • transformations (if used)
  4. Click Save (top-right)
Shared artifact

If others can view the dashboard, they will see your changes after you save.
For experiments, consider duplicating the dashboard or panel first.

Choose a visualization type (panel type)​

Each panel has a visualization type (Grafana calls this a visualization).
You can change it from the panel editor—near the top-right of the editor, below the Save area.

Grafana’s official documentation explains each visualization and its options. Use these links for details:

Core visualizations (common)​

Plugin visualizations (environment-dependent)​

These are common plugins that may be enabled in your Capture environment:

Screenshots (legacy but still useful for orientation):

Best practices (performance and usability)

The following patterns improve dashboard usability and reduce load on the data source—especially for large time ranges.

Use rows to group and collapse sections​

Rows help you structure dashboards into collapsible sections (for example: “Overview”, “Details”, “Diagnostics”).

Benefits:

  • less clutter for users
  • faster dashboards when large sections remain collapsed (depending on panel settings and data source behavior)

How to add a row:

  1. Click Add
  2. Select row
  3. Drag existing panels into the row
Operator-friendly layout

Put “must-see” KPIs above.
Put heavy charts and diagnostic panels in a row that can be collapsed.

Combine multiple fields in a single query (when possible)​

If multiple values come from the same measurement/series, consider retrieving them in one query instead of multiple separate queries. This can significantly reduce query overhead.

Example shown below:

  • three separate queries took 29.86 seconds
  • one combined query took 13.18 seconds

Keep readability

Performance matters, but so does maintainability.
Combine queries when it stays understandable; otherwise document intent in panel descriptions.

Group by time interval using $__interval​

When querying a large time range, returning raw high-frequency data can overwhelm both the data source and the browser. Use time grouping to reduce the number of points rendered.

A common pattern is:

group by time($__interval)

$__interval automatically adapts to:

  • the selected time range
  • the panel width/density

This reduces clutter and improves performance.

Rule of thumb

If a dashboard feels slow over long time ranges, add $__interval grouping first.
Then optimize queries further only if needed.

Next steps​

  • View a dashboard: variables, time ranges, refresh, and “No data” troubleshooting
  • Manage (Rights Management): folder/dashboard permissions and best practices for shared dashboards