Microsoft says new authoring capabilities for Fabric Real-Time Dashboards are generally available, adding Design Tiles, Multi-KPI visuals and expanded KPI styling for dashboard authors. The change affects teams using Real-Time Dashboards to monitor operational data, compare related measures and give viewers more context on the dashboard canvas.
Real-Time Dashboards in Microsoft Fabric are used to monitor, analyse and share streaming and historical data through visual tiles. Most query-backed tiles combine a Kusto Query Language query with a visual representation of the results. The new authoring controls are aimed at the presentation layer, where dashboards often become harder to read as service health, deployment status, business metrics and diagnostic detail accumulate.
Design Tiles add context to the canvas
Design Tiles add three building blocks directly to the Real-Time Dashboard canvas. A Title tile adds a dashboard heading or section label, a Text Box adds descriptions, guidance, annotations and links, and a Divider creates a labelled horizontal break between sections.
Authors add these tiles from the top toolbar in Editing mode. Titles and Text Boxes can be edited directly on the canvas, without opening a separate configuration panel. Title formatting applies to the full title, while Text Box formatting can apply to selected text or to text entered next.
Text Boxes also support links and background formatting. Design Tiles use theme-aware, accessible colours and support keyboard navigation. Pasted text is handled to reduce the risk that unsupported formatting or content disrupts the dashboard.
Multi-KPI reduces repeated tiles
Multi-KPI lets one query produce many KPI items in a single visual. If a query returns labelled rows, each row becomes a KPI and its label value becomes the item label. If a query returns a single row with several numeric columns, each numeric column becomes a KPI and the column name becomes the item label.
This matters for operational dashboards where related measures often come from the same query context. A summary by service can become one KPI per service, while a set of named aggregate columns can become a compact operational summary. Authors do not need to create a separate query and tile for each value in those cases.
The visual uses labels from the query results and applies one shared configuration across all KPI items. Authors choose Number, Bar, Gauge or Donut, then configure value format, scale, thresholds, colours and directionality for the set. They can also control the maximum number of items and the arrangement of the KPIs as the tile is resized.
At smaller tile sizes, each KPI shows its value and status colour, with shortened labels available through a tooltip. With more space, the visual can show full labels and supporting details. The aim is to keep related metrics comparable without filling the dashboard with separate tiles.
KPI styling covers status and targets
The same four KPI styles are available for individual KPI tiles and Multi-KPI visuals. Number is used for a formatted value in dense summary areas. Bar shows progress across a horizontal range, while Gauge and Donut show a value within a range using an arc or ring.
Gauge styling includes 180, 220 and 270 degree arcs, band thickness, flat or rounded cap style and gradient fill. Thresholds define healthy, warning and critical ranges. Directionality controls whether higher or lower values indicate declining performance, such as a higher error rate or lower throughput.
A reference line can show a target, baseline or expected value. Formatting controls cover decimal precision, thousands separators and compact notation such as 1.2K. Used consistently, these settings help viewers understand both the current value and the meaning of its status colour.
Operational dashboards need query and layout discipline
The Fabric documentation positions Real-Time Dashboards as part of monitoring and investigation workflows. Supply chain examples use dashboards for shipment updates and delay alerts. Connected factory examples use them to move from an all-factories view to asset-level detail.
Those scenarios show why the new controls are more than cosmetic. A dashboard can contain accurate visuals and still make viewers work too hard to find the relevant measure, interpret a threshold or move from a summary to supporting detail. Grouping related KPIs and adding context on the canvas reduces that friction when the underlying queries already return suitable results.
The prerequisites for creating a Real-Time Dashboard remain a workspace with Microsoft Fabric-enabled capacity and a KQL database with data. Data sources can include Eventhouse or KQL Database, Ontology, Azure Data Explorer and Azure Monitor. The announcement does not state a separate price for Design Tiles, Multi-KPI or KPI styling.
For teams embedding Real-Time Dashboards in web applications, Microsoft Fabric Embed remains in preview. During preview it supports only Real-Time Dashboards, requires a Fabric capacity F SKU, uses Microsoft Entra ID user sign-in and relies on delegated user authentication. Service-principal authentication and app-owns-data embedding are not supported in that preview.
Where KQL results become operational screens for service health, supply chains or factory telemetry, the choices around KPI grouping, thresholds and dashboard layout are part of ECS's Data topic, with Microsoft Fabric, Power BI, SQL Server and Dataverse. Sessions and workshops give attendees a place to compare modelling choices, Real-Time Dashboard authoring patterns and practitioner experience from Fabric projects. Bring dashboard designs, KQL result shapes and capacity questions to the conference for discussion with peers and speakers.
Photo: Mike Reyher, CC BY 2.0, via Wikimedia Commons, cropped
This article was written by the ECS editorial team with the assistance of Amal Agent, run.events' AI assistant, which supported the research, drafting and proofreading.