Skip to main content
Speckle Intelligence dashboards are available on your plan. For current limits and pricing, see your workspace billing or the Speckle pricing pages.
Dashboards let you create custom, interactive views of your model data. Analyze properties, compare versions, and share insights with your team—all without leaving Speckle.
If you blend model data with other sources, join the Databricks integration study on the User testing opportunities page.
If you are here for validation widgets specifically, use Validation Widgets in Dashboards.

Get started

1

Access Dashboards

From your Speckle workspace, click Intelligence in the top left to open the workspace dashboards page. You can also open dashboards from within a project: open the project, then click the Dashboards tab in the project navigation (with Models, Issues, Automations, Collaborators, Settings).
2

Create your first dashboard

Click Add dashboard to create a new dashboard. You’ll be taken to the dashboard editor.
3

Add a model viewer

Drag a Model Viewer widget from the left sidebar onto the canvas. Select a project and model to display. The model viewer acts as the data source for other widgets.
4

Add analysis widgets

Drag additional widgets (charts, tables, etc.) from the left sidebar. Widgets will automatically display data from the model you’ve added.
5

Configure interactions

Click on widgets to configure their settings. Widgets can filter data, colorize the model, and interact with each other.

Dashboard navigation overview

The images below illustrate the structure and layout of the dashboard.
Speckle intelligence nav 01
Speckle intelligence nav 02

Bring in your data

Your dashboard needs at least one model; the Model Viewer widget provides that data to every other widget. The model viewer displays your 3D model; a Dual viewer widget lets you compare two models with a swipe/slider reveal from one shared camera. In dashboards, each data source has a 1:1 relationship with a model loaded via a model viewer. Model-less data sources are not yet supported.

Adding models

To add a model:
  1. Drag the Model Viewer widget onto the canvas
  2. Select a project from the dropdown
  3. Select a model from that project
  4. Choose whether to load the latest version automatically or pin to a specific version
Use latest if you want the dashboard to update when new versions are published. Pin to a specific version if you need to analyze a particular state of the model. You can add models from any project you have access to within the workspace. Open the Sources sidebar to see mounted models. Sources are grouped by project so it is easier to find the right mount when several projects contribute to one dashboard.

Change a pinned version

When a model has moved on and you want the same charts and tables to follow a different version — without deleting the source, remounting it, and re-pointing every widget — change the pin in place from the Sources sidebar.
  1. Open Sources in the left sidebar.
  2. On the model row, choose Change version (or the pin control that lets you pick another version or latest).
  3. Confirm the new version (or unpin to track latest).
The dashboard saves immediately. Widgets keep their bindings to the same source id and recompute against the new version’s properties. Paths that only existed on the old version go empty until you rebind them. You still cannot mount two sources on the same model and version (or both tracking latest for that model).

Comparing models and versions

Some widgets support multiple data sources. You can compare different versions of the same model, models from different disciplines, or models from different projects. Widgets with multiple sources offer two modes:
  • Aggregate mode: Combines data from all sources into a unified view. Use this to analyze data across multiple models or versions together.
  • Compare mode: Shows differences between sources. Use this to identify what’s changed between versions or what differs between models.
For supported widgets, compare mode can render results side by side so grouped distributions are easier to inspect without switching context.

Model viewer features and sync

The Model Viewer and Dual viewer widgets share the same 3D controls. Hover over a viewer to show the toolbars.

Top-right toolbar

  • Lock interactions: Turn on to prevent pan, zoom, and orbit so viewers can be compared without accidental moves.
  • Save / load camera view: Save the current view so the model loads at that angle next time; clear to reset.
  • Fit: Frame the full model (or current selection) in view.
  • Projection: Switch between perspective and orthographic.

Bottom toolbar

The bottom toolbar (Measure, Section, View modes, Light controls) matches the main Speckle 3D viewer. For details, see Interface and Navigation; for measure, section, and lighting in depth, see Exploration and Presentation.

Sync camera (sync views)

When the dashboard has at least two data sources (e.g. two model viewers, or a Dual viewer with two models), each viewer widget has a Sync camera option in its sidebar. Turn Sync camera on for the viewers you want to keep in sync. When sync is on, panning, zooming, or orbiting one viewer updates the others so you can compare models from the same viewpoint. Sync requires at least two data sources; the option is disabled otherwise. See the video here on how to use sync camera.

Dual viewer

Dual viewer compares two models or versions with a swipe/slider between panes. The cameras in both panes stay synced — you always navigate one shared viewpoint.
  • Both panes need a model before either side renders.
  • Section box and measurements are not available on Dual viewer (they are misleading on a split view).
  • Bottom view-mode controls drive the left pane; modes mirror to the right.
Use Dual viewer when you want a spatial reveal between two schemes. Use two Model Viewer widgets with Sync camera when you prefer independent canvases that stay lined up.

Context models

Use a context model when a model should appear in the 3D view for spatial reference but should not be affected by dashboard filters in that viewer. A site model, existing conditions, or surroundings mesh are typical examples when you are analyzing design schemes on the same canvas. By default, every model loaded in a Model Viewer is a data model. Dashboard filters isolate or hide objects in data models, and colorization from widgets applies to them. To mark a model as context:
  1. Load two or more models in the same Model Viewer (sidebar ModelsAdd model).
  2. Select the viewer and open Models in the widget sidebar.
  3. Open the menu on the model row.
  4. Choose Use as context model.
The model row shows a pinned-map icon (data models show a layers icon). Hover the icon for a short description. Choose Use as data model in the same menu to revert. When a model is a context model:
  • Its geometry stays fully visible when page-level filters isolate objects in your data models.
  • Filters that only target a context model are ignored by that viewer.
  • Widget colorization driven by filters does not isolate or ghost context model objects.
Context models are set per Model Viewer widget. Charts and tables still read from the data sources you connect them to. When a viewer loads several models, connect analysis widgets only to the design data sources if you want counts and breakdowns to exclude a site or reference model.
Yes. Add multiple model viewer widgets; other widgets can then connect to any of them as their data source.
Yes. As long as both models are in Speckle, you can compare them. The comparison works on shared properties between the models.
Yes. If you have access to the project you can add its models to a dashboard.
Dashboard data updates when you load or reload the page. If the model viewer uses “latest”, it fetches the latest version at that time.
Loading or refreshing a dashboard that displays model data counts as a productive receive toward your workspace’s monthly sync usage. So opening a dashboard or reloading the page (when the model viewer fetches a version) consumes sync. For how syncs are defined, how limits work, and how to view usage, see Sync Usage in the New plans FAQ.
Dashboards are defined at the workspace level. If a project or model that a dashboard uses is deleted, the dashboard will break—widgets that depend on that data will no longer have a valid source.
Yes for Scenarios. Apply a scenario and share the dashboard URL with ?scenario=<id> so recipients open the same filters and colour-by. Other configuration (project, model, or arbitrary filter params) is not passed via the URL yet — tell us what you need via the Dashboards category on the Speckle Community or Intercom.
Use a context model when you need geometry in the 3D view (for example a site or existing building) but do not want dashboard filters to isolate or hide it. Typical cases: comparing several design options in one viewer while keeping the site visible, or showing surroundings while charts tabulate only the schemes you connect as data sources.

Widget groups in the sidebar

The widget panel in the dashboard editor groups widgets by category. Each group has a particular focus. Model Validation lives in its own category (Validation) and is shown separately in the sidebar; availability may depend on your workspace plan.
This section is for dashboard-based validation widgets. For that path, use Validation Widgets in Dashboards. For Data Validation (checks and standards), use Data Validation Overview.
The exact groups and widgets you see depend on your workspace and connected data sources. Use the search box in the widget panel to find a widget by name or description. Experimental widgets are marked with a Beta badge in the widget panel so you can quickly identify features that are still evolving.

Choosing the right widget

Use this table to pick a widget for what you want to do. 🔄 = supports Compare mode (multiple data sources).

Aggregation and numeric display

Total property value and Total value by property calculate with the method you pick in widget SettingsAggregation: When Aggregation is Sum, Total property value also shows a secondary average under the primary number. Several numeric widgets also have display controls in Settings:
  • Rounding — Always available. Options: 2 decimals (default), 1 decimal, 0 decimals, Auto, Locale, or Off.
  • Display unit — Appears when the selected property (or ratio operand) has area units. Choose Source, m², ha, ft², or mm². Aggregation stays in source units; the widget converts for display, then rounds. Auto rounding may switch large areas to hectares (for example 12,500 m² → 1.25 ha).
These controls apply to Total property value, Total value by property, Ratio value, Ratio value by property, and Numeric range. On Numeric range, labels use the display settings; filtering and viewer colouring still use source units.
Widget Display unit only changes how numbers appear. To convert units for calculated fields, filters, or colour-by, create a Unit conversion calculated field instead.

How to connect a widget to a model

How to connect a widget to multiple models

Partially. You can customize dashboards with built-in Text / Markdown, Image, and Embed widgets, and with Themes (chart and validation colors). There is not yet a public API to build custom widgets that plug into the dashboard. If you need custom widgets or integrations, get in touch via the Dashboards category on the Speckle Community or Intercom. For custom applications built on dashboard capabilities (Enterprise), see the next FAQ.
Yes. Enterprise customers have been working with us to deliver custom applications via dashboard capabilities. Build with Speckle describes how we partner to extend and embed Speckle. If you want custom dashboards, integrations, or embedded analytics, get in touch via the Dashboards category on the Speckle Community or Intercom.

Chart types

Many widgets (Count by property, Total value by property, and others) let you switch between chart and list views. Use the Chart Style or Type setting in the widget to choose how data is shown.
Some chart widgets let you adjust margins and label placement. Select the widget, open its Settings in the sidebar, and look for options such as margins, padding, or label position. Increase margins or change the label setting so labels fit inside the widget. If the widget has no label options, try resizing the widget or switching to a different chart type (e.g. Grid or List) to see values in full.
Yes, for widgets that support multiple chart types. Use the Chart Style or Type setting in the widget settings to switch between Pie, Bar, Tree, Grid, List, or Drop down. Some widgets (e.g. time-based or version-over-time) use a fixed chart type and do not offer this switcher.
Long labels or many categories can overflow the chart area. If the widget supports it, adjust margins or label settings in the widget settings. Otherwise, resize the widget, reduce the number of categories (e.g. with a filter), or switch to Grid or List to see all values in a table.

Property selection and the property library

You can choose which properties to display in two ways: inside a widget, or from the property library. You also select properties when adding filters (e.g. filter by Category or Level); see Filters for filter levels and operators.
Many widgets (Count by property, Total property value, Element table, and others) prompt you to select one or more properties for display. Use the widget’s settings or property dropdown to pick the property; the widget then shows values, counts, or breakdowns for that property.es. How to Add an Element Table with Multiple Properties

Explore and filter

Filters let you focus on specific parts of your data. Chart and table widgets drive page-level filters (chips at the top that affect all widgets) and optional widget-level filters (local to one widget). You can also colorize the 3D viewer from widget data. Operators depend on the property’s data type (text, number, list, boolean); filters combine as OR within one chip and AND across chips. For full detail on filter levels, operators, properties, and colorizing the model, see Filters. To add your own values from model data without changing the published model, use Calculated fields. To import a reusable list of codes and labels for those fields (or for validation), use Lookup tables. When a widget value points back to a model data source, you can open that source in the viewer directly from the dashboard flow to inspect context without leaving your analysis.

How to apply multiple filters using charts

How to color code and change color themes

Organize and share

You can resize and reorder widgets, add Section widgets to group areas, and use the Themes panel for chart and validation colors. Dashboards are shared at the workspace level; roles control who can view or edit. Duplicate a dashboard to use it as a template. For layout and themes, see Layout. For roles, share links, and presentation mode, see Sharing. For named filter and colour-by snapshots in edit and presentation mode — including deep links — see Scenarios on the Sharing page. Presentation mode is what anyone you share with always sees (also called view mode or shared mode). For step-by-step guides (quantity takeoff, portfolio analysis, finding a family), see Common workflows.

Getting help

For questions about dashboards, visit the Dashboards category on the Speckle Community.
Last modified on July 18, 2026