How to Get Started with Microsoft Fabric

Microsoft Fabric is Microsoft's unified analytics platform delivered as a SaaS product. Business users who already work in Excel or Power BI can connect data sources, run transformations, and publish reports inside a single environment without managing servers, databases, or infrastructure. A free 60-day trial gives full access to all platform features. This guide covers the key components you need to know and the four practical steps to run your first report.

What Microsoft Fabric Is and Why Analysts Should Care

Microsoft launched Fabric for general availability in November 2023 as a consolidation play: instead of paying for separate tools for data storage, data transformation, and business intelligence, organizations get all three bundled under one licensed platform. The underlying architecture uses a storage layer called OneLake, which is built on Azure Data Lake Storage and serves as the single location where all data across all workloads is kept. Analysts do not need to know the technical details to use the platform effectively, but understanding that all data lives in one place explains why sharing a dataset between two teams in Fabric is a click rather than a project.

According to Hex's 2026 State of Data Teams report, AI tooling increased 24% as a focus area for data teams compared to the previous year, and data trust was cited as the top concern around AI adoption. Fabric's Copilot features, which sit inside the Power BI and Data Factory interfaces, reflect exactly this trend: AI assistance for generating reports and transformations, applied on top of data that the organization governs centrally rather than scattered across files and spreadsheets.

The Workloads You Actually Need at the Start

Fabric is organized into workloads, each designed for a different role. At the start, business analysts and operations managers need three of them.

Power BI is the reporting and visualization workload. It is the same tool as Power BI Service and Power BI Desktop but embedded inside Fabric. Reports built in standalone Power BI Desktop can be published to a Fabric workspace without modification. This is the primary output layer for most business users.

Data Factory is the data ingestion and transformation workload. It uses a Power Query interface similar to the one inside Excel, and supports over 200 connectors to databases, cloud services, spreadsheets, SharePoint lists, and APIs. Analysts who have built Power Query transformations in Excel will recognize every step.

OneLake is the storage layer underneath everything. Every file, table, and dataset created inside a Fabric workspace lands in OneLake automatically. The practical benefit is that a dataset loaded once can be used in multiple reports by multiple teams without creating duplicate copies or maintaining manual export processes.

The remaining workloads, including Data Engineering, Data Science, Real-Time Intelligence, and Data Warehouse, are relevant for data engineers and technical teams. A business analyst does not need them to run reports.

Starting the Free Trial

Microsoft offers a 60-day free trial of Fabric at app.fabric.microsoft.com. Signing in requires a Microsoft account or a work email. An Azure subscription is not required; Fabric is a standalone SaaS product.

When you sign up, Microsoft provisions a trial capacity automatically. Capacity determines how much compute is available for processing queries and running transformations. The trial capacity handles individual use and small teams exploring the platform comfortably.

If your organization uses Microsoft 365, your IT administrator may have already enabled Fabric access. In that case, you can open app.fabric.microsoft.com directly and start working without activating a separate trial.

Creating a Workspace

A workspace in Fabric is a container for all data, reports, pipelines, and other artifacts belonging to a team or project. It functions like a shared folder with access controls.

To create one, open Fabric and click "Workspaces" in the left navigation panel, then select "New workspace." Enter a name that reflects the team or project, for example "Finance Reporting" or "Operations Analytics." You can add colleagues at this step and set their permission level to Viewer, Contributor, or Admin.

Once created, the workspace shows an empty home screen. The most common starting items for business analysts are a Dataflow Gen2 (for loading and transforming data using Power Query) and a Report (for building Power BI visualizations).

Connecting a Data Source with Dataflow Gen2

Dataflow Gen2 is the recommended way to bring external data into a Fabric workspace for reporting. It uses the Power Query editor and stores the output in OneLake.

From the workspace, click "New item" and choose "Dataflow Gen2." The Power Query editor opens. Click "Get data" to see the list of available connectors. Common options for business users include:

• Excel workbook or CSV file uploaded from your computer
• SharePoint Online list or folder
• SQL Server or Azure SQL Database
• Salesforce, Dynamics 365, or Google Analytics

After connecting to a source, Power Query loads a preview of the data. Apply any needed transformations directly in the editor, such as removing blank rows, changing column data types, renaming headers, or filtering to a specific date range. When you click "Publish," Fabric runs the transformation and writes the result to OneLake as a table.

Dataflows support scheduled refresh. Set a daily or hourly refresh schedule from the dataflow settings page and Fabric will re-run the transformation automatically, keeping the data current without manual intervention.

Building a Report in Power BI

With data in your workspace, create a Power BI report on top of it. From the workspace, click "New item" and choose "Report." The Power BI editor opens in the browser.

In the "Data" panel on the right side, all tables available in your workspace are listed. Drag a field onto the canvas to place a visual. Power BI suggests a chart type based on the data fields you select. You can switch chart types from the Visualizations panel, adjust colors and labels, and add report-level filters using the "Filters" pane.

Copilot in Fabric appears as a panel in the Power BI editor. Type a request in plain language, such as "show monthly revenue by product category as a bar chart," and Copilot generates the visual and adds it to the canvas. The output usually needs formatting adjustments, but Copilot produces a working starting point for most standard chart types.

When the report is ready, click "Save" and then "Share" to generate a link. Anyone with the link and appropriate workspace permissions can view the report in a browser without installing software.

What to Expect on Performance and Cost

The free trial runs on trial capacity, which is shared and may respond more slowly than a paid tier. For a team running ad hoc reports on datasets under a few million rows, trial performance is usually adequate. Processing very large datasets or running complex transformations may be noticeably slower.

After the trial, Fabric pricing is based on capacity units. The smallest paid tier, F2, costs approximately $262 per month as of early 2026 and is appropriate for a small team running reports on business-scale data volumes. Organizations that already license Power BI Premium P1 or higher have access to Fabric capabilities included in their existing subscription.

For analysts who work primarily with file uploads rather than live database connections, VSLZ handles the upload, transformation, and chart generation from a single prompt with no workspace or capacity to configure.

Next Steps

The practical first sequence for a new Fabric user is straightforward: sign up for the trial, create a workspace, connect one data source with Dataflow Gen2, and publish one Power BI report. That sequence typically takes under an hour and covers the full workflow from raw data to a shareable report. The free Fabric Analyst in a Day workshop, available at Microsoft Learn, provides hands-on exercises that extend this foundation to more complex scenarios including lakehouse queries and multi-source data modeling.