Microsoft Power BI Desktop

Description

Business data lives in dozens of places at once — in an ERP system, in a CRM database, in Excel files on a shared drive, in a cloud data warehouse, in a web analytics platform, in a SQL Server instance, in an accounting system that outputs only to CSV. Turning that scattered data into a coherent picture of how the business is doing used to require a data engineering team, weeks of work and expensive enterprise software. Power BI Desktop provides a single analyst the means to connect to all of those sources, clean and model the data, define business metrics, and create interactive dashboards — all in a free Windows application that publishes results to the web with a Power BI Pro subscription.

Power BI was released as the companion desktop application to the Power BI cloud service by Microsoft in July 2015 as Power BI Desktop. It is at the authoring end of the Power BI platform: analysts create reports in the desktop application and publish them to the Power BI Service, where they can be viewed and interacted with by their colleagues using a browser or the Power BI mobile app. The desktop application is updated on a monthly basis, with Microsoft releasing a new version of the application on the last Tuesday of each month, which incorporates new visuals, data connector support, and modeling capabilities.

KEY FEATURES

Data Connectivity

Power BI Desktop connects to over 100 data sources with a connector library spanning databases (SQL Server, Oracle, MySQL, PostgreSQL, Snowflake, Google BigQuery), cloud (Azure, Salesforce, Dynamics 365, SharePoint, Google Analytics), files (Excel, CSV, JSON, XML, PDF) and Web data using the Web connector that can scrape the data from an HTML table on public URLs. Each connector manages authentication separately — service account credentials for databases, OAuth for cloud services and API keys where needed. Multiple data sources are connected simultaneously in a single report and data from different sources is combined using the data model.

Power Query Editor

The Power Query Editor is a tool for transforming raw data from your source into analysis-ready tables without writing SQL or code. Transformations can be applied using a graphical interface: renaming columns, changing data types, filtering rows, pivoting and unpivoting table structures, merging tables from different sources, splitting columns using delimiter, and replacing values. Every transformation step is recorded in a list called Applied Steps, which displays the complete history of the transformation and makes it possible to modify or remove any step without starting over. Power Query generates the M language code behind the interface, and for advanced users, they can write M language directly for transformations that are not covered by the graphical tools.

Data Modeling and DAX

The data model describes relationships between tables — linking a sales transactions table with a products table using a product ID, or linking a customers table with an orders table using a customer ID. Relationships allow for cross-table filtering: when you select an area of a map visual, every other visual on the page is filtered to display only values pertaining to that area. DAX (Data Analysis Expressions) refers to the formula language of creating calculated columns and measures. Measures define business metrics — total revenue, year over year growth percentage, running totals, moving averages — as formulas that recalculate based dynamically on the current filter context in any visual. DAX is used to perform complex time intelligence calculations such as period comparisons and cumulative totals using a library of built-in time intelligence functions.

Report Canvas and Visuals

The report canvas contains the visuals that comprise a report page. Built-in visual types include bar and column charts, line charts, scatter charts, pie and donut charts, treemaps, maps, tables, matrices, KPI cards, gauges, and slicers. Each visual binds to fields from the data model — to drag a date field to the axis of a line chart and a revenue measure to the values creates a revenue over time chart, with no further configuration. Custom visuals from the AppSource marketplace include chart types that are not available as defaults, such as specialized statistical charts, advanced maps, and other custom KPI visualizations that are created by third parties. All visuals on a page cross-filter each other by default — click a bar in one chart and every other visual on the page will be filtered to the selected dimension.

Row-Level Security

Row-level security limits which rows of data specific users can see when they access a published report. A report on regional sales, for example, may limit each sales manager to viewing the data for his or her own region, even though the underlying data include all the regions. Security roles are used to define filter rules in DAX and users or groups are assigned to roles in the Power BI Service after the publishing.

Publishing and Sharing

Completed reports are published to the Power BI Service with the click of a button from the desktop application and become immediately available in the browser to users with Power BI Pro licenses. Published reports are embedded in SharePoint pages, Teams tabs, and external websites using iframe embed codes. The Power BI mobile app for iOS and Android shows published reports in a touch optimized view.

SYSTEM REQUIREMENTS

• Operating system: Windows 8.1 or newer (64-bit) Windows 10 or 11 recommended
• Memory: 2 GB RAM minimum; 8 GB recommended for large data sets
• Storage: 2.84 GB free disk space
• Display: 1440 x 900 or better resolution is recommended
• Internet: Necessary for Publishing to Power BI Service and Online Data Source Access
• Note: Power BI Desktop is Free – Sharing reports with other users requires Power BI Pro licenses or Premium capacity