Power BI Introduces Real-Time Streaming Semantic Models

Microsoft has announced a game-changing feature for Power BI: streaming semantic models that deliver real-time data to dashboards without refresh delays.

The Problem with Scheduled Refreshes

Until now, Power BI relied on scheduled refreshes — meaning data could be up to 30 minutes or more out of date. For operations teams, finance, and customer support, this lag meant decisions based on stale data.

How Streaming Semantic Models Work

The new feature connects directly to data sources via Event Hubs, Kafka, or Dataverse real-time APIs and streams changes into the semantic model continuously.

Key capabilities:

Sub-second latency from source to dashboard
Automatic aggregation of streaming data into time windows
Hybrid mode — combine streaming data with historical batch data in the same model
Built-in alerting when KPIs cross thresholds

Use Cases

Manufacturing — Monitor production line output in real time
E-commerce — Track orders, revenue, and cart abandonment live
Customer support — Watch ticket volume and response times as they happen
Finance — Real-time cash flow and transaction monitoring

Requirements

Power BI Premium or Premium Per User (PPU) license
Azure Event Hub or compatible streaming source
Dataverse real-time sync (for Power Platform data)

Availability

Rolling out in public preview now. GA expected June 2026.

Source: Microsoft Power BI Blog — March 25, 2026