Juggle Dash guide

Dashboard as code, without the provisioning

“Dashboards as code” often means provisioning: store a dashboard definition in Git, then apply it to a running service through a deployment process. Juggle Dash does not provision dashboards or apply configuration to a server. It exports a JSON representation of an editable dashboard draft, so you can inspect and keep the definition without a server or credentials.

What the dashboard JSON contains

The DSL JSON export brings together the dataset fields, the dashboard's semantic panels, and their positions on the canvas. A field records a name, an inferred type such as date, string, or number, and a role such as dimension or measure. Nodes describe panels such as KPI cards, trend lines, category bars, and ranking tables. Each node also carries a frame describing its placement and size on the 12-column dashboard canvas.

In a SaaS revenue example, month can be a date dimension, channel a string dimension, and mrr a numeric measure. Those fields can support a monthly MRR trend or comparisons such as channel ROI and ARR by segment. A customer dimension can be used to present a top-customer ranking. The data rows and panel configuration travel with the exported definition.

A trimmed excerpt from the SaaS revenue export

{
  "version": 1,
  "theme": "enterprise_blue",
  "canvas": { "width": 1920, "height": 1080, "gridCols": 12 },
  "meta": { "title": "SaaS Revenue Operations", "source": "csv" },
  "data": [{
    "id": "saas-revenue-operations",
    "fields": [
      { "name": "month", "type": "date", "role": "dimension" },
      { "name": "channel", "type": "string", "role": "dimension" },
      { "name": "mrr", "type": "number", "role": "measure" }
    ]
  }],
  "header": { "title": "SaaS Revenue Operations" },
  "nodes": [{
    "id": "mrr-trend",
    "type": "trend_line",
    "title": "MRR Trend",
    "frame": { "x": 32, "y": 316, "w": 916, "h": 320 }
  }]
}

This shortened excerpt shows the actual DSL's version, theme, canvas, field roles, header, trend panel, and frame shape. It leaves out the full rows and chart options to keep the example readable; it is not a complete importable dashboard file on its own.

Semantic panels, not just coordinates

The definition records more than a collection of pixel positions. A trend panel has a panel type and chart configuration, while the frame places that panel in the dashboard grid. The generation workflow proposes KPI and chart meaning, and deterministic layout rules assign the panel frames. That separation gives a reviewer both the intent of a panel and its visual placement in one structured file.

Because the export is JSON, it can be opened in an editor, searched, or compared in a version-control diff. Diffs are most useful when you keep the same dashboard and compare deliberate revisions; regenerating from changing input data can naturally produce broader changes in embedded rows or chart values.

The real SaaS revenue operations export illustrates that combination: its dataset lists the dimensions and measures, its nodes identify KPI cards, a trend line, category charts, and a ranking table, and every node has a frame on the shared canvas. The chart panels contain their chart options as JSON as well, so the file carries their display configuration alongside the field names and values. The exported structure is therefore useful for understanding how a particular draft was assembled, not merely for recording where boxes appeared on screen.

Keeping the definition with your project

Export DSL JSON from the editor when you want a copy of the definition for inspection, review, or project records. You can continue refining the active draft in the visual editor and export the updated JSON again. The current editor workflow does not import an exported DSL JSON file back into a new session, so do not treat the JSON file as a supported round-trip restore format. For a view that can be opened directly in a browser, export HTML instead.

Read when to choose the HTML snapshot if your main goal is delivery. If you have not created a dashboard yet, the CSV import guide explains how source fields become a draft. The product overview also describes the features and visual editor.

For a practical review, open the JSON in a text editor and locate the top-level canvas, dataset, header, and node sections. Check that the selected field names match the source columns, the panel titles explain their measures, and the frame values place each panel on the canvas. A reviewer can then discuss the shape and intent of a dashboard without mistaking the file for a running service or a configured data connection.

What this format is not

Juggle Dash DSL JSON is not a BI semantic layer, a Terraform provider, or a provisioning API. It does not contain database connection strings, a credentials vault, or a live source connection. The exported data is a snapshot of the dataset used for the draft, and a definition does not refresh itself as source data changes.

This distinction matters when researching a dashboard JSON format. If you need a declarative artifact to deploy dashboards into a particular platform, use that platform's own provisioning schema and workflow. Juggle Dash's JSON is for the dashboard definition it creates and exports.

Export a definition from a real draft

Start with a CSV or JSON file, use a prompt or English preset to produce a first draft, then arrange the panels and export DSL JSON from the editor. Open the editor to see the workflow. For plan details, visit Juggle Dash pricing.