Skip to content

Power BI

Power BI is Microsoft's business intelligence platform: Power BI Desktop to model data and build reports, the Power BI service to share and govern them, and mobile apps to view them. It is also a core workload of Microsoft Fabric, and capacity licensing is moving from Power BI Premium P SKUs to Fabric F SKUs.1

Intended for
Business users who read reports, Report creators and analysts, BI and IT administrators, Developers who embed analytics, Decision makers evaluating BI
Runs on
Power BI service (SaaS), Power BI Desktop on Windows, Power BI Mobile apps, Power BI Report Server on premises, Embedded in custom applications

At a glance

What is it?
Power BI is a business analytics platform for connecting to data, modeling it and publishing interactive reports and dashboards. Its main parts are Power BI Desktop (a Windows app for modeling with Power Query and DAX), the Power BI service (cloud workspaces, apps, sharing, refresh and security), the Power BI Mobile apps, Power BI Report Builder for paginated, print-ready reports and Power BI Report Server for keeping reports on premises. Microsoft describes Power BI as a core component of Microsoft Fabric that shares OneLake, Direct Lake mode, Copilot and Purview-based governance with the other Fabric workloads.1
What does it do?
Authors connect Power BI Desktop to more than 100 data sources, shape data in Power Query, build models with relationships and DAX measures, and design reports. Publishing to the service lets teams share through workspaces and apps, schedule refresh, set alerts and email subscriptions, apply row-level security and sensitivity labels, embed content and review usage metrics and audit logs. Inside Fabric, semantic models can read OneLake Delta tables in Direct Lake mode instead of importing data.1
Who needs it?
Business analysts and report authors who turn data into reports, staff and managers who read them, and BI or IT teams who administer workspaces, capacities and security. Developers use Power BI embedding to put reports into internal or customer-facing applications. Organizations still on Power BI Premium P SKUs need to plan their move to Fabric F SKUs.1
What does it need?2
  • A work or school account in a Microsoft Entra tenant
  • Power BI Desktop (a Windows app) for report authoring
  • A Power BI Pro or Premium Per User license for each person who creates or shares content in shared workspaces
  • For large audiences of readers: a Fabric F64 or larger capacity
  • For OneLake and Direct Lake: a Fabric capacity and data stored in OneLake
What it is not
Power BI is not the whole of Microsoft Fabric: Pro and Premium Per User licenses do not provision Fabric capacity, so lakehouses, warehouses and notebooks still need an F capacity. Power BI Desktop is the authoring tool, not the sharing platform, and sharing outside My workspace needs Pro or Premium Per User. Premium Per User is not being retired; only the per-capacity P SKUs are. Power BI is also not a data warehouse: large-scale storage and transformation belong in the database, lakehouse or warehouse it connects to.23

Availability and licensing. Generally available. Microsoft's pricing page (checked 26 Oct 2026) lists Power BI Pro at $14.00 and Premium Per User at $24.00 per user per month, paid yearly, and notes that shown prices are for marketing and the checkout price applies. Power BI Premium per-capacity P SKUs are no longer sold; each retires at the end of its agreement term, then gets a 30-day grace period, throttling from day 31 and rejected operations from day 91. Microsoft directs customers to Fabric F SKUs. Premium Per User is not retiring.34

The problem it solves

Spreadsheet reporting does not scale well: figures differ between files, refreshes are manual and access is hard to control. Teams need one shared model of their metrics, reports that refresh on a schedule and a controlled way to share them.

Power BI separates authoring (Desktop) from distribution (the service), lets one semantic model feed many reports, and applies row-level security, sensitivity labels, usage metrics and audit logs in the service.1

How it works

A typical Power BI setup has four parts:

  • Power BI Desktop: connect to sources, transform with Power Query, model with DAX and design reports.
  • Power BI service: workspaces, apps, scheduled refresh, subscriptions, alerts, row-level security, sensitivity labels, usage metrics and audit logs. Workspaces run on shared capacity (Pro or PPU workspace types) or on a Fabric F capacity (or a legacy P capacity).
  • Semantic models: import, DirectQuery or, on Fabric, Direct Lake over Delta tables in OneLake.
  • Consumption: browser, Power BI Mobile apps, embedding, Excel and Teams through Microsoft 365 integration, and paginated reports built in Report Builder.

Licensing decides who can do what: on F64 and larger capacities, users with a free license and the Viewer role can read content; below F64, every viewer needs Pro or PPU.25

Power BI architecture: components by layer and how they connectCopilot & businessappsData & analyticsDeveloper toolsSecurity, identity &governanceCloud & hybridinfrastructurePower BI service: Workspaces, apps, refresh, subscriptions and alertsPower BI serviceBrowser, mobile apps, Excel, Teams and embedded apps: Where readers consume reportsBrowser, mobile apps, Excel,Teams and embedded appsData sources (files, databases, cloud services, OneLake): Where the data livesData sources (files,databases, cloud services,…Semantic model (import, DirectQuery or Direct Lake): Shared definition of tables, relationships and measuresSemantic model (import,DirectQuery or Direct Lake)Power BI Desktop: Authoring tool for Power Query transforms, DAX models and report designPower BI DesktopRow-level security, sensitivity labels and audit: Access control and compliance, shared with Fabric through PurviewRow-level security,sensitivity labels and auditShared capacity or Fabric F capacity: Compute that runs refreshes and queries; decides viewer licensingShared capacity or Fabric Fcapacity
Diagram as a list
  1. Copilot & business apps

    • Power BI serviceWorkspaces, apps, refresh, subscriptions and alerts
    • Browser, mobile apps, Excel, Teams and embedded appsWhere readers consume reportsConnects to Power BI service
  2. Data & analytics

    • Data sources (files, databases, cloud services, OneLake)Where the data lives
    • Semantic model (import, DirectQuery or Direct Lake)Shared definition of tables, relationships and measuresConnects to Data sources (files, databases, cloud services, OneLake)
  3. Developer tools

    • Power BI DesktopAuthoring tool for Power Query transforms, DAX models and report design
  4. Security, identity & governance

    • Row-level security, sensitivity labels and auditAccess control and compliance, shared with Fabric through Purview
  5. Cloud & hybrid infrastructure

    • Shared capacity or Fabric F capacityCompute that runs refreshes and queries; decides viewer licensing
Components and connections as documented by Microsoft.

Capabilities

  • Power BI Desktop modeling12

    Connects to more than 100 sources, transforms data in Power Query Editor, builds models with DAX, calculated columns and relationships, and supports composite models and the performance analyzer.

    Why it matters: One well-built model can serve many reports and users.

    Limits: Windows app. Semantic models that use more than 1 GB of memory need Premium Per User or a capacity.

  • Sharing and governance in the service1

    Workspaces, apps, scheduled refresh, email subscriptions, alerts, row-level security, sensitivity labels, usage metrics and audit logs.

    Why it matters: Reports reach the right people with access rules applied centrally.

    Limits: Creating or sharing Power BI items outside My workspace requires Pro or Premium Per User.

  • Direct Lake on OneLake13

    Semantic models query Delta tables in OneLake without importing or caching the data.

    Why it matters: Reports on large Fabric datasets stay current without scheduled import refreshes.

    Limits: Needs data in OneLake and a Fabric capacity; Direct Lake guardrails such as row counts, file counts and model size scale with capacity size.

  • Paginated reports and Report Server13

    Report Builder creates fixed-layout, printable reports such as invoices; Power BI Report Server keeps reports on premises.

    Why it matters: Covers operational and regulatory reporting that needs exact page layouts.

    Limits: After moving from P SKUs, Report Server rights come with a Fabric capacity reservation or SQL Server Enterprise Edition with Software Assurance.

  • Embedding23

    Embed for your organization (Microsoft Entra users) or for customers (app owns data, authenticated by a service principal). Power BI Embedded is included with every F SKU.

    Why it matters: Puts analytics inside existing applications and portals.

    Limits: On F SKUs below F64 and on A SKUs, each internal viewer needs Pro, Premium Per User or an individual trial.

  • Copilot and natural language Q&A1

    Copilot helps create reports and find insights, and Q&A lets users ask questions of data in natural language.

    Why it matters: Lowers the barrier for readers who do not know the model.

    Limits: Answers depend on a clearly named, well-modeled semantic model; review AI-generated content before relying on it.

  • Per-user and capacity licensing23

    Pro and Premium Per User are per-user licenses; Fabric F capacities license by compute. Premium Per User adds 48 refreshes a day, models over 1 GB and the XMLA endpoint.

    Why it matters: Lets organizations choose per-user licensing for small groups and capacity for broad audiences.

    Limits: Power BI Premium P SKUs are no longer sold and retire at the end of each agreement term.

Practical use cases

Finance sends monthly results as spreadsheets and different teams quote different figures.
Approach
Build one semantic model in Power BI Desktop, publish it to a workspace and distribute reports through an app with row-level security.
Role of Power BI
Power BI holds the shared model, the reports and the access rules.
Data, infrastructure and skills
Pro or PPU licenses for authors (and for viewers below F64), and access to the finance data sources.
Type of benefit
Consistent figures and controlled access
Caveats
Model quality decides report quality; plan refresh windows and security roles early.
First step
Rebuild one recurring finance report in Power BI Desktop and publish it to a test workspace.
An organization on Power BI Premium P1 faces the renewal of its Enterprise Agreement.
Approach
Buy an F64 capacity in Azure first, reassign and validate workspaces, then cancel the P SKU.
Role of Power BI
Power BI content moves unchanged to a Fabric F capacity.
Data, infrastructure and skills
An Azure subscription, capacity admins and a usage baseline from the Capacity Metrics app.
Type of benefit
Continuity of service
Caveats
Autoscale does not exist on F SKUs; cross-region moves do not carry large semantic models or Fabric items.
First step
Review 30 to 45 days of P SKU usage in the Microsoft Fabric Capacity Metrics app to size the F SKU.
A software company wants dashboards inside its customer portal without giving customers Power BI accounts.
Approach
Use the app-owns-data embedding pattern with a service principal on an F SKU.
Role of Power BI
Power BI renders the reports; the application handles user sign-in.
Data, infrastructure and skills
An F, A or P capacity, a service principal and application code using the embedding APIs.
Type of benefit
New product capability
Caveats
Row-level security must be passed from the app correctly; capacity size limits concurrent load.
First step
Embed one report in a test page using a small F SKU.

Who uses it

Works with

Optional integration

  • Azure DatabricksPower BI Desktop connects to Databricks SQL warehouses, and data can be published from Databricks to the Power BI service.
  • Azure Synapse AnalyticsAzure Synapse lists deep integration with Power BI.

Complementary tools

Same family

Optional integration for

Relationship labels follow Microsoft's documentation. "Alternative approaches" does not mean one is better: each profile says when it fits.

Compared with

Getting started

  1. Install and connect

    Download Power BI Desktop, sign in with a work account and connect to a source such as Excel or a SQL database.

    Check: The data appears in the Power Query Editor.

  2. Model and build

    Clean the data, create relationships and DAX measures, and build a report page.

    Check: Visuals respond correctly to filters and slicers.

  3. Publish and share

    Publish to a workspace in the Power BI service and share through an app.

    Check: A colleague with the right license and role can open the report.

  4. Refresh and secure

    Set scheduled refresh, add row-level security roles and apply sensitivity labels.

    Check: Test users see only the rows their role allows.

  5. Choose licensing

    Compare Pro, Premium Per User and an F capacity against your number of authors and readers.

    Check: Every reader is covered by a license or by an F64 or larger capacity.

Official resources

Deciding whether you need it?

Three free tools apply Microsoft's own published guidance to your situation, and say when Power BI or any new technology is not needed.

Sources

Each statement above links to the source it comes from. Labels say who reported it.

  1. What is Power BI? (opens in a new tab) Microsoft · Vendor-reported
  2. Understand Microsoft Fabric licenses and capacity (opens in a new tab) Microsoft · Vendor-reported · link checked 27 Oct 2026
  3. Power BI Premium to Microsoft Fabric migration FAQ (opens in a new tab) Microsoft · Vendor-reported
  4. Power BI pricing (opens in a new tab) Microsoft · Vendor-reported
  5. What is Microsoft Fabric (opens in a new tab) Microsoft · Vendor-reported · link checked 27 Oct 2026

Fill out the form below to request your copy.

Name(Required)