Recent from talks
Nothing was collected or created yet.
Microsoft Power BI
View on WikipediaThis article needs to be updated. (June 2023) |
| Microsoft Power BI | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Developer | Microsoft | ||||||||
| Initial release | July 11, 2011 | ||||||||
| Stable release(s) [±] | |||||||||
| |||||||||
| Operating system | Windows 10, 11 iOS 17, 18[b] watchOS 2+ Android 8+ web | ||||||||
| Type | Data visualization Business intelligence | ||||||||
| License | Proprietary | ||||||||
| Website | powerbi | ||||||||
Microsoft Power BI (PBI) is an interactive data visualization software product developed by Microsoft with a primary focus on business intelligence (BI).[8] It is part of the Microsoft Power Platform. Power BI is a collection of software services, apps, and connectors that work together to turn various sources of data into static and interactive data visualizations. Data may be input by reading directly from a database, webpage, PDF, or structured files such as spreadsheets, CSV, XML, JSON,[9] XLSX, and SharePoint.[10]
General
[edit]Power BI provides cloud-based BI (business intelligence) services, known as "Power BI Services", along with a desktop-based interface, called "Power BI Desktop". It provides data warehouse capabilities including data preparation, data mining, and interactive dashboards.[11] In March 2016, Microsoft released an additional service called Power BI Embedded on its Azure cloud platform.[12] One main differentiator of the product is the ability to load custom visualizations.
History
[edit]The software was originally used as Power Pivot and Power Query in Microsoft Excel. This application was originally conceived by Thierry D'Hers and Amir Netz of the SQL Server Reporting Services team at Microsoft.[13] It was originally designed by Ron George in the summer of 2010 and named Project Crescent.[14] Project Crescent was initially available for public download on 11 July 2011, bundled with SQL Server Codename Denali.[15] Later renamed Power BI it was then unveiled by Microsoft in September 2013 as Power BI for Office 365.[16] The first release of Power BI was based on the Microsoft Excel-based add-ins: Power Query, Power Pivot and Power View. With time, Microsoft also added many additional features like question and answers, enterprise-level data connectivity, and security options via Power BI Gateways.[11] Power BI was first released to the general public on 24 July 2015. It has several versions for desktop, web, and mobile app.[17]
On 14 April 2015, Microsoft announced that they had acquired the Canadian company Datazen, to "complement Power BI, our cloud-based business analytics service, rounding out our mobile capabilities for customers who need a mobile BI solution implemented on-premises and optimized for SQL Server." [18] Most of the 'visuals' in Power BI started life as Datazen visuals.
Key components
[edit]
Key components of the Power BI ecosystem are as follows:
- Power BI Desktop[19]
- The Windows desktop-based application for PCs, primarily for designing and publishing reports to the service.
- Power BI Service
- The SaaS-based (software as a service) online service. This was formerly known as Power BI for Office 365, now referred to as PowerBI.com or simply Power BI.
- Power BI Mobile Apps
- Power BI Mobile apps are for Android and iOS devices, as well as for Windows phones and tablets.
- Power BI Gateway
- Gateways are used to sync external data in and out of Power BI and are required for automated refreshes. In enterprise mode, it can also be used by Microsoft Power Automate (previously called Flows) and PowerApps in Office 365.
- Power BI Embedded
- Power BI REST API can be used to build dashboards and reports into the custom applications that serve Power BI users and non-Power BI users.
- Power BI Report Server
- An on-premises Power BI is a reporting product for companies that choose not to store data in the cloud-based Power BI Service.
- Power BI Premium
- Capacity-based offering that includes flexibility to publish reports broadly across an enterprise without requiring recipients to be licensed individually per user. This provides greater scale and performance than shared capacity in the Power BI Service.
- Power BI Visuals Marketplace
- A marketplace of custom visuals and R-powered visuals.[20]
- Power BI Dataflow[21]
- A Power Query implementation in the cloud that can be used for data transformations to make a common Power BI Semantic Model, which can then be made available for report developers through Microsoft's Common Data Service. For example, it can be used as an alternative to doing transformations in SSAS and may ensure that several report developers use data that has been transformed similarly.
- Power BI Semantic Model[22]
- A Power BI Semantic Model (formerly Dataset) can work as a collection of data for use in Power BI reports, and can either be connected to or imported into a Power BI Report.[23] A semantic model can be connected to and get its source data through one or more dataflows.
- Power BI Datamart
- Within Power BI, the datamart is a container that combines Power BI Dataflows, datasets, and a type of data mart or data warehouse (in the form of an Azure SQL Database) into the same interface. The interface then has the possibility of being a single place for the administration of both the ETL layer (Dataflow), an intermediary data mart (with for instance storage of star schemas, dimension tables, fact tables), and finally the modeling layer (dataset).
- Power BI Datahub
- A data hub for discovering Power BI datasets within an organization's Power BI Service so that datasets may be reused from one central location. It offers details on the things as well as an access point for working with them, such as building reports on top of them, utilizing them with Excel's Analyze feature, accessing settings, controlling permissions, and more.[24]
Power Query
[edit]ETL processes in both the web and desktop versions of Power BI are facilitated in Power Query through built-in connectors to pull data from a wide variety of sources. Power Query provides a GUI which allows users to perform many common data preparation operations without needing to write code, though more advanced operations may be performed through expressions written in the M formula language.[25]
Licenses
[edit]Power BI has many different licenses depending on use.
User licenses
[edit]Some of the most common user licenses are free, pro, and premium. Free users must be part of an organization with a Power BI license. They can consume reports, and can also build, but not publish. In practice, creators need at least a pro license ($14.00/month) in order to publish reports, and premium users ($24.00/month) can publish like a pro but also have more features available for developing.[26]
Workspace licenses
[edit]Premium capacity refers to the license of the workspace, not the licenses of its users or creators. Also, access to several features, functionalities, and kinds of content that are exclusively accessible through premium is made possible with a Power BI Premium per-user license.[27] Premium per user is a special workspace license which was added in November 2020,[28] and can be a more affordable alternative to premium workspace licenses for organizations with few users which have advanced analytical requirements.[28]
Paginated reports
[edit]Paginated reports for Power BI, which can be built with Power BI Report Builder, are a special type of SSRS reports with pagination formatting which can give better control of the layout of reports which need to be printed to paper or PDF. This is in contrast to regular Power BI reports which instead are optimized for presentation or interactivity and exploration on a screen. Paginated reports can, as of 2022, not be made with the regular Power BI Desktop report builder software. Instead, the standalone Power BI Report Builder has to be used, which can be viewed as a descendant of the SQL Server Reporting Services (SSRS) Microsoft Report Builder for Microsoft SQL Server introduced in 2004. It is also similar to the Report Designer in SQL Server Data Tools.
Power BI Paginated reports are saved in the Report Definition Language (.rdl file format), as opposed to the .pbix file of regular Power BI reports or the recently introduced .pbir format.[29] The RDL format is based on XML and was proposed by Microsoft as a benchmark for defining reports with SSRS.
Paginated reports may be more suitable than regular Power BI reports, and may include printing of invoices or other repeated printouts of reports with a similar layout but different content, or for printing reports where text would otherwise overflow due to being cut off by scrollbars.
Notes
[edit]References
[edit]- ^ "Power BI Desktop". Microsoft Apps. Retrieved 20 January 2026.
- ^ "Microsoft Power BI Desktop". Microsoft. Retrieved 20 January 2026.
- ^ Murray, Katie (20 January 2026). "Power BI January 2026 Feature Summary". Microsoft Power BI. Retrieved 20 January 2026.
- ^ "Microsoft Power BI". App Store. Retrieved 20 January 2026.
- ^ "Microsoft Power BI". Google Play. Retrieved 20 January 2026.
- ^ "Microsoft Power BI 2.2.251204.21280310". APKMirror. 4 December 2025. Retrieved 20 January 2026.
- ^ "Power BI". Microsoft Marketplace. Retrieved 20 January 2026.
- ^ "Bring your data to life with Microsoft Power BI". Microsoft.com. Microsoft. Archived from the original on 27 May 2019. Retrieved 4 May 2016.
- ^ "List of all Power Query connectors - Power Query". learn.microsoft.com. 31 March 2023. Archived from the original on 11 July 2023. Retrieved 11 July 2023.
- ^ "How to Design Power BI Dashboard: A Step-by-Step Guide". lookuplinks.com. 5 August 2023. Archived from the original on 26 September 2023. Retrieved 24 September 2023.
- ^ a b "Magic Quadrant for Business Intelligence and Analytics Platforms". Gartner.com. Gartner, Inc. Archived from the original on 29 May 2018. Retrieved 4 May 2016.
- ^ Caldwell, Nick (31 March 2016). "Embed the wow of Power BI in your applications with Microsoft Power BI Embedded!". Powerbi.microsoft.com. Microsoft. Archived from the original on 18 February 2019. Retrieved 4 May 2016.
- ^ "A Glimpse at Project Crescent". Microsoft BI Blog. Microsoft. 9 November 2010. Archived from the original on 24 March 2016. Retrieved 29 June 2017.
- ^ "Data Visualization Done Right: Project Crescent". Microsoft BI Blog. Microsoft. 9 November 2010. Archived from the original on 31 August 2017. Retrieved 29 June 2017.
- ^ Unkroth, Kay (12 July 2011). "Announcing Microsoft SQL Server Code Name "Denali" Community Technology Preview 3 (CTP3)". blogs.msdn.microsoft.com. Microsoft. Archived from the original on 25 April 2017. Retrieved 28 May 2019.
- ^ "Office 365 Gets Colorful 3D Charts, Natural Language Search". PC Mag India. ZiffDavis, LLC PCMag India. 26 September 2013. Archived from the original on 16 December 2019. Retrieved 4 May 2016.
- ^ The Power BI Team, Microsoft (10 July 2015). "Announcing Power BI general availability coming July 24th". Microsoft Power BI Blog. Microsoft. Archived from the original on 12 August 2021. Retrieved 4 May 2016.
- ^ Hathi, Kamal (14 April 2015). "Microsoft acquires mobile business intelligence leader Datazen". blogs.microsoft.com. Microsoft, Inc. Archived from the original on 10 January 2022. Retrieved 10 January 2022.
- ^ Hyman, Jack (8 February 2022). Microsoft Power BI for dummies. NJ: wiley. pp. 47–62. ISBN 978-1-119-82487-9.
- ^ "Power BI Custom visuals on AppSource". Microsoft AppSource. Microsoft. 25 December 2017. Archived from the original on 26 December 2017. Retrieved 25 December 2017.
- ^ "Creating a dataflow - Power BI | Microsoft Docs". Archived from the original on 24 February 2022. Retrieved 24 February 2022.
- ^ "Datasets renamed to semantic models". powerbi.microsoft.com. Retrieved 3 July 2025.
- ^ Get started, Power BI (3 October 2022). "Basic concepts for designers in the Power BI service". Archived from the original on 24 March 2022. Retrieved 4 April 2022.
- ^ paulinbar. "Data discovery using the data hub - Power BI". learn.microsoft.com. Archived from the original on 11 March 2023. Retrieved 11 March 2023.
- ^ Puls, Ken; Escobar, Miguel (2021). Master Your Data with Excel and Power BI: Leveraging Power Query to Get & Transform Your Task Flow ; Formerly M is for Data Monkey (2nd ed.). Chicago: Holy Macro! Books. ISBN 978-1-61547-058-7.
- ^ "Power BI: Pricing Plan | Microsoft Power Platform". www.microsoft.com. Retrieved 9 September 2025.
- ^ kfollis. "Power BI service features by license type - Power BI". learn.microsoft.com. Archived from the original on 27 November 2022. Retrieved 27 November 2022.
- ^ a b Meet Power BI Premium per user element61
- ^ "Power BI's Enhanced Report Format (PBIR) - Acuity Training". www.acuitytraining.co.uk. Retrieved 9 September 2025.
Further reading
[edit]- Gunnarsson, Ásgeir; Johnson, Michael (2020). Pro Microsoft Power BI Administration: Creating a Consistent, Compliant, and Secure Corporate Platform for Business Intelligence. Apress. ISBN 978-1484265666.
External links
[edit]Microsoft Power BI
View on GrokipediaOverview
Definition and Purpose
Microsoft Power BI is a unified, scalable platform for self-service and enterprise business intelligence (BI), serving as Microsoft's interactive data visualization software product that enables users to connect to, visualize, and share data from various sources.[9] As a core component of the Microsoft Power Platform, it focuses on transforming raw data—such as from databases, Excel files, PDFs, and web services—into interactive reports, dashboards, and actionable insights to support informed decision-making across organizations.[1] The primary purpose of Power BI is to democratize data access and analysis, allowing non-technical users to explore and interpret complex datasets without requiring advanced coding skills, thereby fostering a data-driven culture in businesses.[9] It delivers real-time analytics capabilities, enabling timely insights through seamless integration with tools like Microsoft Fabric, and offers scalability to handle enterprise-level data volumes while maintaining security and governance.[1] These features make it particularly effective for creating visually compelling representations of data trends and patterns. High-level use cases for Power BI include tracking sales performance to identify top-performing products and regions, financial forecasting to predict revenue and budget variances, and operational monitoring to optimize supply chain efficiency and resource allocation.[9] By embedding analytics into the broader Microsoft ecosystem, such as Microsoft 365, it enhances productivity for teams relying on collaborative decision-making.[1]Architecture and Deployment Options
Microsoft Power BI employs a client-server architecture that separates content authoring from consumption and collaboration. At the client layer, Power BI Desktop serves as the primary tool for designing reports and semantic models, allowing users to connect to data sources, transform data, and create interactive visualizations locally before publishing. The server layer, hosted in the cloud via the Power BI Service, handles sharing, collaboration, and advanced analytics, while mobile apps enable on-the-go consumption across iOS and Android devices. This model ensures scalability and accessibility, with data processed through Azure backend services for storage and computation.[11][12] The data flow in Power BI begins with ingestion from diverse sources, including databases, files, and web services, often facilitated by tools like Azure Data Factory for enterprise-scale pipelines. Ingested data is then modeled into semantic models—optimized datasets that define relationships, measures, and hierarchies—to support efficient querying. These models power visualization creation, where reports and dashboards render interactive elements like charts and maps. Refresh mechanisms maintain data currency: scheduled imports update datasets periodically, while DirectQuery mode queries live sources in real-time, and incremental refresh optimizes large datasets by updating only changed partitions.[12][11] Deployment options for Power BI cater to varying organizational needs, balancing cloud convenience with on-premises control. The cloud-based SaaS model via the Power BI Service offers fully managed hosting on Azure, with automatic scaling, monthly feature updates, and seamless integration for collaboration. For on-premises scenarios, Power BI Report Server provides a self-hosted alternative, deployable on local servers or virtual machines, supporting paginated reports and import modes but lacking real-time streaming and requiring manual updates three times annually. Embedded deployment leverages Power BI Embedded in Azure, allowing developers to integrate analytics into custom applications using APIs for authentication and rendering, with capacities billed hourly and support for service principal identities.[13][14] Security is embedded in Power BI's architecture through row-level security (RLS), which filters data visibility based on user roles defined in semantic models, applicable in both import and DirectQuery modes across deployment options. Data governance is enforced via workspace permissions, sensitivity labels, and audit logs in the Power BI Service, while on-premises setups rely on SQL Server integration for similar controls. Hybrid environments use on-premises data gateways to securely bridge local sources to cloud resources without exposing credentials.[13][11][12]History
Origins and Early Development
The origins of Microsoft Power BI trace back to 2011, when it was conceived as Project Crescent by a team within Microsoft's SQL Server Reporting Services group, led by Thierry D'Hers and Amir Netz.[15][16] This initiative aimed to create an immersive, ad-hoc visualization tool integrated with SQL Server "Denali" (later SQL Server 2012), emphasizing self-service business intelligence accessible to non-technical users.[17] Project Crescent built directly on foundational Microsoft technologies, including Power Pivot—a columnar data compression engine (VertiPaq) introduced as an Excel add-in in 2010 for in-memory analytics—and Power Query, an ETL tool for data transformation also originating from Excel extensions.[18][19] These components extended the capabilities of earlier enterprise tools, allowing users to model and analyze large datasets without heavy IT involvement. Power BI's conceptual roots extended further to Microsoft's longstanding business intelligence ecosystem, particularly SQL Server Analysis Services (SSAS), which debuted in 1998 as part of SQL Server 7.0's OLAP Services and evolved into a core engine for multidimensional and tabular data modeling. By the early 2010s, SSAS had matured into a robust platform for semantic data layers, providing the analytical foundation that Project Crescent leveraged to enable interactive reporting and exploration.[15] This heritage ensured Power BI inherited proven scalability for handling complex queries and aggregations, while shifting focus toward cloud-native, user-friendly interfaces to democratize BI beyond traditional on-premises deployments. Early development progressed through previews that highlighted Power BI's cloud-based ambitions. In July 2013, Microsoft launched a public preview of Power BI for Office 365, rebranding Project Crescent and introducing browser-based data discovery, natural language querying, and integration with Excel for seamless self-service analytics.[20] This phase emphasized lightweight, web-accessible tools for visualizing data from diverse sources, setting the stage for broader adoption. A pivotal enhancement occurred in April 2015, when Microsoft acquired Datazen Software, a mobile BI specialist, to integrate its high-performance dashboards and cross-platform visualization capabilities into Power BI.[21] The acquisition addressed gaps in mobile responsiveness and real-time rendering, accelerating Power BI's evolution into a comprehensive, device-agnostic platform.Key Milestones and Evolution
Microsoft Power BI reached general availability on July 24, 2015, marking its transition to a standalone business intelligence platform after being renamed from the earlier Project Crescent initiative.[4][22] This launch enabled broader access to interactive data visualization and reporting tools, initially attracting 500,000 early adopters across 45,000 companies.[23] Subsequent releases expanded Power BI's capabilities and distribution. In September 2017, Power BI Desktop became available through the Microsoft Store, simplifying installation and updates for users on Windows devices.[24] Power BI Premium entered general availability in June 2017, introducing dedicated capacity options to support larger-scale deployments and advanced governance features without per-user licensing constraints.[25] In 2019, the platform integrated AI-powered visuals, such as Key Influencers for identifying factors affecting demand, Decomposition Tree for hierarchical analysis, and Anomaly Detection for spotting irregularities in sales data, to automate insight discovery in datasets.[26][27] Power BI Premium Per User licensing preview began in November 2020, with general availability following in April 2021, offering individual access to premium features like AI visuals and larger dataset handling.[28] From 2023 onward, Power BI evolved through deeper integration with Microsoft's ecosystem, particularly with the launch of Microsoft Fabric in May 2023 and its general availability in November 2023, unifying data engineering, science, and analytics workflows while embedding Power BI for seamless reporting.[29] Monthly feature updates continued to drive innovation, including native ARM support for Power BI Desktop in the September 2025 release to optimize performance on Windows on ARM devices.[30] The October 2025 update enhanced Copilot capabilities, enabling general availability of DAX query generation and editing through natural language prompts in both Desktop and web environments.[31] The November 2025 update introduced general availability for semantic model version history (retaining up to 5 previous versions) and enhancements to the card visual, alongside preview features like standalone Copilot in mobile apps.[32] By July 2025, Power BI marked its 10-year anniversary, having grown from its initial adopter base to serve over 375,000 organizations worldwide, including 95% of the Fortune 500, and processing vast datasets that underscore its role in enterprise-scale analytics.[23][33] This evolution reflects Power BI's shift from a visualization tool to a comprehensive analytics platform, consistently ranked as a leader in the industry.[23]Core Components
Power BI Desktop
Power BI Desktop is a free Windows application designed for report authoring and data modeling, enabling users to connect to various data sources, transform data, create semantic models, and design interactive visualizations within .pbix files.[6] It serves as the primary tool for developers and analysts to build comprehensive reports offline, supporting end-to-end creation from data import to visual layout without requiring an internet connection during authoring.[34] A core functionality of Power BI Desktop is the creation of semantic models, which organize data into relationships, hierarchies, and calculations to provide a unified view for analysis. Users leverage Data Analysis Expressions (DAX), a formula language comprising over 250 functions, to define measures and calculated columns that enable advanced computations such as time intelligence and aggregations.[35][36] For visualization, the application offers a drag-and-drop canvas with hundreds of built-in and partner-provided visuals, alongside an open-source framework for developing custom visuals to meet specific needs. Power BI Desktop supports adding images to reports that can be configured with interactive actions, such as page navigation to other report pages, enhancing user navigation in multi-page reports. Detailed steps for this feature are covered in the Reporting and Analytics Features section.[37][38] Offline authoring allows full report development without connectivity, while scheduled data refreshes can be configured to update datasets automatically upon publishing.[34] System requirements for Power BI Desktop include Windows 10 or later (or Windows Server 2016 or later), .NET Framework 4.7.2 or later, at least 2 GB of RAM (4 GB recommended), a 1 GHz 64-bit processor, and a display resolution of 1440x900 or higher.[34] As of the October 2025 update, it provides native support for Windows on ARM PCs with the 2025-09 cumulative update (KB5065789), enhancing performance and battery efficiency on compatible devices.[30] Reports created in Power BI Desktop can be exported to PDF for static sharing or to Excel/CSV for data extraction from visuals, and published directly to the Power BI Service for broader distribution.[39][40]Power BI Service and Mobile Apps
The Power BI Service is a cloud-based Software as a Service (SaaS) platform that enables users to publish reports and dashboards created in Power BI Desktop, facilitating collaboration and consumption of data insights across organizations.[1] It supports the creation of workspaces, which act as shared environments for teams to manage and collaborate on content, from which apps can be published to package and distribute curated reports and dashboards to end-users.[11] Key functionalities include real-time data alerts that notify users of important changes in metrics via email or the service interface, and content endorsement features such as promotion and certification for datasets, reports, dashboards, and dataflows to increase visibility and trust. Promotion adds a "Promoted" badge, improves search ranking, and is available to workspace Admins, Members, and Contributors (those with Contribute permission) who have edit or build access to the content. Certification adds a "Certified" badge for official content, is restricted to specific users or security groups designated by the Power BI tenant admin, and is disabled by default at the tenant level.[41] These elements ensure secure sharing while enforcing data governance, with the service integrating seamlessly with Microsoft Fabric for broader analytics workflows.[42] Complementing the service, Power BI offers native mobile applications for iOS, Android, and Windows devices, providing on-the-go access to reports and dashboards without requiring full authoring capabilities.[43] Users can view interactive content, add annotations to highlight insights during reviews, and receive push notifications for data alerts or updates directly on their devices, enhancing responsiveness to business needs.[44] The apps support offline access to cached reports for scenarios with limited connectivity, allowing users to explore visuals and refresh data upon reconnection, while maintaining security through integration with the service's authentication.[45] Collaboration within the Power BI Service is bolstered by features like row-level security (RLS), which dynamically filters data based on user roles to ensure individuals see only relevant information, enforced across shared reports and in mobile views as well.[46] Additionally, the built-in Q&A natural language querying tool allows users to ask questions in plain English about their data, generating instant visualizations or answers from underlying semantic models, promoting self-service analytics without technical expertise.[47] These capabilities extend to workspace roles, which control permissions for viewing, editing, and administration to foster secure team interactions. Workspace roles include: Admin, providing full control (add/remove members, manage settings, promote/certify content if allowed); Member, allowing collaboration, editing, and promotion of content; Contributor, enabling creation and editing of content along with promotion; and Viewer, restricted to read-only access with no ability to promote or certify. These endorsement and workspace management features are covered in the PL-300: Microsoft Power BI Data Analyst exam under managing workspaces and endorsing content.[48] In 2025, the Power BI Service introduced enhanced Copilot features, including a generally available standalone full-screen experience for analyzing reports and data via natural language prompts directly in the cloud environment, improving accessibility for non-technical users.[7]Power Query for Data Preparation
Power Query serves as the primary data preparation tool within Microsoft Power BI, functioning as an ETL (Extract, Transform, Load) engine that enables users to ingest, clean, and reshape data from diverse sources before loading it into semantic models for analysis.[18] It features a graphical user interface for intuitive operations alongside the M formula language for scripting complex transformations, allowing for repeatable and automated data workflows.[49] Integrated across Power BI Desktop, Power BI Service (via dataflows), Excel, and other Microsoft tools like Azure Data Factory, Power Query ensures consistent data handling regardless of the platform.[18] The tool supports connectivity to over 140 data sources through built-in connectors, spanning categories such as databases (e.g., SQL Server, Oracle), cloud services (e.g., Azure Blob Storage, Google BigQuery), file formats (e.g., Excel, CSV, JSON), and business applications (e.g., Salesforce, Dynamics 365).[50] Key processes include extracting data via these connectors, transforming it through cleaning operations like removing duplicates or handling null values, and combining datasets by merging (joining tables on common keys) or appending (stacking rows from similar tables).[51] Users can also incorporate parameters for dynamic queries—such as filtering by date ranges—and define custom functions in M to encapsulate reusable logic, facilitating scalable data preparation for large-scale BI projects.[49] Advanced capabilities enhance performance and extensibility, including query folding, which pushes transformation steps back to the source database for execution, minimizing data transfer and improving refresh times—particularly beneficial for large datasets where supported operations like filtering or grouping are translated into native SQL statements.[52] Additionally, the Power Query SDK allows developers to create custom connectors for proprietary or niche data sources, using Visual Studio Code to build, test, and deploy extensions in formats like .mez files, thereby expanding integration beyond standard options.[53] The output of Power Query transformations results in structured datasets that are loaded directly into Power BI's semantic models, serving as the foundation for subsequent modeling, visualization, and reporting without requiring external ETL tools.[51]Reporting and Analytics Features
Interactive Visualizations and Dashboards
Microsoft Power BI enables the creation of interactive visualizations that transform prepared data into dynamic, user-engaging displays, allowing users to explore insights through intuitive interactions. These visuals are built using Power BI Desktop or the service, drawing from semantic models connected via tools like Power Query for data preparation. Core visual types include a variety of charts such as bar, column, line, pie, donut, area, scatter, ribbon, waterfall, funnel, and combo charts, which support filtering and highlighting to reveal patterns in datasets.[54] Maps, including basic, filled (choropleth), ArcGIS, Azure, and shape maps, facilitate geographic analysis by plotting data on interactive globes or regions.[54] Key performance indicators (KPIs) provide at-a-glance metrics with conditional formatting for targets and trends, while slicers act as interactive filters, enabling users to segment data by categories like date, region, or product without altering the underlying dataset.[54] Power BI provides built-in time series forecasting for line chart visuals through the Analytics pane, allowing users to predict future values based on historical data using exponential smoothing (ETS/Holt-Winters algorithms). Users can adjust parameters such as forecast length and confidence intervals, with seasonality handled automatically or through algorithmic detection. Advanced forecasting can be achieved by integrating R or Python scripts for custom models like SARIMA. This supports applications such as demand forecasting in retail and consumer packaged goods, where dashboards can incorporate factors like promotions, multi-channel data, and hierarchical forecasts.[55][56] Power BI's AppSource marketplace extends these options with certified custom visuals developed by partners, allowing tailored representations for specific needs. Examples include timeline slicers, which offer compact, scrollable date-range selections for efficient time-based filtering, and advanced decomposition trees that enable hierarchical data exploration beyond the built-in version.[57] The built-in decomposition tree visual supports multi-dimensional drilling, automatically aggregating data and permitting users to break down metrics across fields like sales by region, product, and time.[58] As of November 2025, the decomposition tree visual was enhanced to support all-expanding nodes with legend field integration for improved customization in hierarchical analysis.[32] Dashboards in Power BI aggregate these visuals into cohesive, single-page overviews for monitoring key metrics and trends. Users build dashboards by pinning individual visuals or entire report pages from published reports in the Power BI service; to pin, one selects a visual in edit mode, clicks the pin icon, and chooses an existing or new dashboard, creating a static or live tile that updates with data refreshes.[59] Interactions enhance usability: cross-filtering automatically adjusts other visuals on the page when a data point is selected, such as clicking a bar in a chart to filter a linked map; drill-through provides deeper navigation by right-clicking a visual to access a dedicated report page filtered to that context, supporting buttons for explicit drill-through paths.[60][61] Power BI also enables page navigation using images as interactive elements, allowing users to create custom links between report pages, such as placing an image on a report's home page that links to another page. In Power BI Desktop, users insert an image via Insert > Image, select the image on the canvas, and in the Format pane, enable the Action toggle, set the Type to Page navigation, and choose the target page from the dropdown list. This configuration makes the image clickable for navigation. To test in Power BI Desktop, hold Ctrl and click the image; after publishing the report to the Power BI service, end users can navigate by directly clicking the image.[38] Best practices for interactive visualizations and dashboards emphasize accessibility, consistent theming, and performance, particularly with large datasets. For accessibility, adhere to Web Content Accessibility Guidelines (WCAG) by adding descriptive alt text (up to 250 characters) to visuals for screen readers, ensuring 4.5:1 color contrast ratios, using shapes or patterns alongside colors for data distinction, and setting logical tab orders via the Selection pane to aid keyboard navigation.[62] Theming maintains visual consistency through JSON-based report themes applied in Power BI Desktop, which propagate corporate colors, fonts, and effects across all visuals; best practices include importing custom JSON files for branding alignment and testing in the service to verify rendering.[63] For performance optimization with large datasets, limit visuals per page to six or fewer, apply restrictive filters like "Top N" to reduce query loads, use import mode where possible for faster rendering, and monitor via Performance Analyzer in Desktop to identify bottlenecks in custom visuals or complex interactions.[64] The key influencers visual uses machine learning (via ML.NET) to automatically analyze and rank factors impacting a selected metric, such as identifying drivers of customer churn through logistic or linear regression on up to 10,000 sampled data points.[65] It improves interpretability with dynamic segment analysis and p-value filtering (threshold 0.05), enabling faster insight discovery in interactive reports and dashboards without manual modeling.[65]Paginated Reports
Paginated reports in Microsoft Power BI are specialized reports designed for precise, print-ready outputs that display all data across multiple pages without requiring scrolling, making them ideal for scenarios demanding fixed layouts and high-fidelity rendering.[66] These reports are authored using Power BI Report Builder, a standalone tool that enables the creation of .rdl files, which are report definition files compatible with the Power BI service and SQL Server Reporting Services (SSRS).[67] Power BI Report Builder supports advanced features such as parameters for user-driven filtering, subreports for embedding nested reports, and expressions for dynamic calculations and formatting, allowing designers to build complex, pixel-perfect layouts including tables, matrices, charts, and free-form elements.[67] Common use cases for paginated reports include generating invoices, financial statements, and compliance reports, where the need for structured, printable documents with exact page control is paramount, such as rendering hundreds of rows across multiple pages without truncation.[68] Unlike interactive Power BI reports, which emphasize dynamic exploration through scrolling and modern visuals, paginated reports prioritize pagination optimized for printing, ensuring all data is presented in a fixed, non-interactive format suitable for PDF exports or physical distribution.[69] Key features of paginated reports extend to robust export capabilities, supporting formats like PDF, Excel, Word, CSV, XML, and MHTML, which facilitate seamless sharing and archiving of detailed operational data.[68] Bursting functionality allows for automated distribution, such as scheduling email subscriptions with PDF attachments on a daily, weekly, or hourly basis, enhancing efficiency in report dissemination.[66] Additionally, paginated reports integrate closely with SSRS, enabling migration of existing .rdl files from SQL Server 2016, 2017, or 2022 using tools like the RDL Migration Tool, while maintaining backward compatibility for on-premises deployments via Power BI Report Server.[68] Once created, these reports can be published directly to the Power BI service for viewing and management.[67]AI and Copilot Capabilities
Microsoft Power BI incorporates artificial intelligence through specialized visuals that automate the discovery of patterns and insights in data. The anomaly detection visual provides built-in anomaly detection for line chart visuals with time series data. It automatically identifies unexpected spikes or dips (anomalies), highlights them, displays the expected value range, and offers natural language explanations analyzing contributing factors from the data model. The feature is enabled via the Analytics pane in the visual's settings. It uses the SR-CNN algorithm, requires at least four data points, and needs a Pro or Premium license. It is supported in Power BI Desktop and the Power BI service. Key limitations include no support for legends, multiple values, secondary values, compatibility with forecast, average, min, max, median, or percentile lines, or certain data sources such as Direct Query over SAP. For advanced batch processing, integration with the Azure Anomaly Detector API is possible.[8][70][71] The Key Influencers visual identifies the key factors that most impact a selected metric, such as factors affecting demand in sales data, by analyzing relationships and ranking drivers of the metric.[27] Power BI provides built-in time series forecasting in line chart visuals through the Analytics pane, using exponential smoothing algorithms to predict future trends based on historical data. This feature automatically handles seasonality, generates confidence intervals, and allows users to adjust parameters such as forecast length. It is particularly applicable to demand forecasting scenarios, such as predicting sales or inventory needs in retail and consumer packaged goods contexts. For advanced forecasting needs, Power BI supports integration with automated machine learning (AutoML) and Azure Machine Learning, as well as custom models using R or Python scripts for techniques like regression or SARIMA.[55][27] The decomposition tree visual facilitates hierarchical data exploration, allowing drill-down across multiple dimensions to uncover relationships and drivers of metrics through an interactive, tree-like interface.[58] Copilot, Microsoft's generative AI assistant, integrates deeply with Power BI to enhance natural language interactions and automate report creation. It supports natural language report generation by converting user prompts into visuals, narratives, and full reports, streamlining the design process for non-technical users.[7] Q&A capabilities are improved through Copilot's semantic understanding, enabling more accurate responses to conversational queries by incorporating synonyms, context, and model metadata.[72] Additionally, Copilot provides summarization features that generate concise overviews of reports, extracting key trends and insights from visuals in seconds.[73] This integration is primarily enabled within Microsoft Fabric, with 2025 updates introducing a standalone Copilot experience for broader accessibility outside the Fabric environment.[7] As of November 2025, standalone Copilot became available in Power BI mobile apps (in preview), allowing chat-based insights and visual generation on the go, while Report Copilot was enhanced for smarter visual creation and an expanded library of supported visuals.[32] In 2025, key advancements include Copilot's capabilities for semantic model tuning, which evaluate and optimize data models by suggesting improvements in relationships, measures, and naming conventions to enhance AI performance and query accuracy.[74] It also delivers real-time insights through integration with Fabric's Real-Time Intelligence, allowing dynamic analysis of streaming data for immediate decision-making.[75] Ethical considerations in Power BI's AI features emphasize responsible use, particularly in detecting and mitigating bias in outputs to ensure fairness and reliability. Microsoft applies its Responsible AI principles, including tools for transparency and bias assessment in models, to address potential disparities in anomaly detection or forecasting results arising from skewed training data.[76] Users are encouraged to review AI-generated explanations and validate insights against diverse datasets to minimize ethical risks.[77]Licensing and Pricing
User Licenses
Microsoft Power BI offers three primary per-user licensing tiers designed for individual and team-based usage: Free, Pro, and Premium Per User (PPU). These licenses determine the scope of data creation, sharing, collaboration, and advanced analytics capabilities available to users within the Power BI service. As of November 2025, pricing and features reflect updates implemented in April 2025, with Free providing basic access at no cost, Pro enabling collaborative workflows for $14 per user per month, and PPU unlocking enterprise-grade features for $24 per user per month (billed annually).[78][79] The Free license, also known as Fabric Free, allows users to create and view reports and dashboards for personal use without any subscription cost. It supports up to 8 data refreshes per day and 10 GB of storage per license, with 1 GB of model memory per dataset. However, sharing and collaboration are strictly limited to personal consumption only; users cannot publish or share content with others unless the workspace is backed by Premium capacity. This tier is ideal for beginners or individual exploration but requires upgrading for team interactions.[78][79] Power BI Pro extends functionality for collaborative environments at $14 per user per month. It enables users to publish reports to the Power BI service, share dashboards and datasets with other Pro license holders, and collaborate in workspaces. Key limits include 1 GB of model memory per dataset, 8 data refreshes per day, and 10 GB of storage per license. Pro users can also consume content shared from Premium workspaces, making it suitable for small to medium teams focused on interactive reporting and basic analytics. An add-on option is available for users with existing Microsoft 365 E5 or Office 365 E5 licenses.[78][79] For advanced needs, Power BI Premium Per User (PPU) provides comprehensive features at $24 per user per month, building on Pro capabilities with Premium-level performance. This includes support for datasets up to 100 GB in model memory, 48 data refreshes per day, and 100 TB of storage, along with AI-powered visuals, paginated reports, and enhanced dataflows. Sharing is restricted to other PPU users, though Pro or Free users can access PPU content if hosted in a Premium workspace. However, Pro users cannot access PPU workspaces directly or utilize PPU-exclusive features such as 100 GB models or advanced AI like AutoML; basic sharing is possible in standard (non-PPU) workspaces, though access to Premium features may be restricted for them. Note that Copilot requires Premium capacity and is not available with PPU licenses. PPU is particularly valuable for power users requiring larger-scale data handling and AI integrations without organizational capacity commitments. A $14 per user per month add-on is offered for Pro or E5 license holders to upgrade to PPU features.[78][79][80][7]| License Tier | Pricing (per user/month, annual) | Key Features | Sharing/Collaboration | Dataset Limits | Refresh Rate |
|---|---|---|---|---|---|
| Free (Fabric Free) | $0 | Personal report creation and viewing; 1 GB model memory | Personal use only; no sharing unless in Premium workspace | 10 GB storage | 8/day |
| Pro | $14 | Publish, share, and collaborate; workspace access | With other Pro users; Free/PPU if in Premium workspace | 1 GB model memory; 10 GB storage | 8/day |
| Premium Per User (PPU) | $24 | All Pro + AI visuals, paginated reports, dataflows; 100 GB model memory | With PPU users; Pro/Free if in Premium workspace | 100 GB model memory; 100 TB storage | 48/day |
Capacity and Embedded Options
Microsoft Power BI offers capacity-based licensing options designed for organizations requiring scalable compute resources beyond individual user licenses. These options provide dedicated processing power measured in virtual cores (v-cores) or capacity units (CUs), enabling features such as large-scale data refresh, advanced AI capabilities, and broad content sharing without per-user licensing for viewers.[82] Premium capacity, now integrated into Microsoft Fabric capacities, is a subscription model that allocates dedicated compute resources for Power BI workloads. It supports up to 100 TB of storage, 48 dataset refreshes per day, and 100 GB per model in memory, facilitating paginated reports at scale and seamless integration with Microsoft Fabric for unified data management across analytics, data engineering, and AI tasks. Fabric capacities use F SKUs, with F64 and larger allowing content consumption by free users in an "app owns data" scenario, eliminating the need for Pro licenses for viewers. Pricing is pay-as-you-go or reservation-based, starting at approximately $8,410 per month for the F64 SKU, which provides 64 CUs equivalent to 8 v-cores for Power BI operations.[83][84][82]| SKU | Capacity Units (CUs) | v-cores (Power BI) | Pay-as-you-go Monthly Price (USD) | Reservation Monthly Price (USD, ~41% savings) |
|---|---|---|---|---|
| F64 | 64 | 8 | $8,409.60 | $5,005.00 |
| F128 | 128 | 16 | $16,819.20 | $10,010.00 |
| F256 | 256 | 32 | $33,638.40 | $20,020.00 |
| Node Type | v-cores | RAM | Hourly Rate (USD) | Estimated Monthly (730 hours, USD) |
|---|---|---|---|---|
| A1 | 1 | 3 GB | $1.00 | $735.91 |
| A2 | 2 | 5 GB | $2.00 | $1,465.91 |
| A3 | 4 | 10 GB | $4.00 | $2,937.67 |
| A4 | 8 | 25 GB | $8.00 | $5,881.25 |
| A5 | 16 | 50 GB | $16.00 | $11,768.33 |
| A6 | 32 | 100 GB | $32.00 | $23,542.94 |
