Microsoft Copilot, Fabric & Power BI: AI-Powered Analytics

Name: Microsoft Copilot, Fabric & Power BI: AI-Powered Analytics
Rating: 4.3 (44 reviews)

Use Copilot, Agents & Power BI MCP Server to Develop Dashboards, DAX & Semantic Models

Created byDr. Florian Detzel | BI Academy

Last updated 6/2026

English

What you'll learn

Use Microsoft Copilot to explore datasets, summarize information, and “talk” to your data through natural language.
Generate DAX, SQL, and PySpark code automatically and refine it with AI-assisted suggestions.
Build Power BI visuals and full dashboard pages using conversational prompts and AI-driven insights.
Understand your data model with Copilot: identify relationships, propose star schemas, and detect anomalies.
Accelerate analytics workflows in Microsoft Fabric using Copilot inside Notebooks for cleaning, analysis, and documentation.
Create AI-powered narratives, summaries, and explanations to enhance storytelling and communication in your reports.
Use Power BI MCP Server to interact with your semantic model from VS Code or Claude AI

Course content

12 sections • 44 lectures • 6h 25m total length

Welcome and getting to know your lecturer3:31
Florian introduces himself, blending theoretical business intelligence and data analysis with practical Power BI experience to deliver solid foundations and hands-on exercises, inviting student feedback.
Course structure and learning objectives6:37
Learn to use Copilot in Microsoft Fabric and Power BI to load data, build semantic models and dashboards, and generate executive summaries and analytics insights for business questions.

Create your Azure subscription1:17
Introduction & create the Azure Resource Group for the Fabric Capacity6:56
Create an Azure resource group and Fabric capacity to enable Copilot in Microsoft Fabric, using pay-as-you-go pricing and Azure credits.
Creating the Microsoft Fabric Capacity9:05
Create a Microsoft Fabric capacity in the Azure portal by selecting the F2 size, configuring your subscription and region, and using pause and resume to manage costs while enabling Copilot.
Creating a capacity workspace in Microsoft Fabric2:40
Create a capacity workspace in Microsoft Fabric, assign the correct fabric capacity named Easycar, and enable Copilot to search reports, answer data questions, and summarize reports in Power BI.

Introducing the sample data2:30
Explore the Easycar sample data with a four-table star schema—reservations fact table plus customer, car, and branch dimensions—and use Copilot in Microsoft Fabric and Power BI to build dashboards.
Download the sample data0:01
Creating a Lakehouse & Upload our data source5:31
Set up a lakehouse in the fabric workspace, upload four csv files, and load them into tables (dim_branches and a fact table) with semicolon separators for Power BI semantic model.
Creating the semantic model for the report in Microsoft Power BI4:25
Create the semantic model in Power BI by connecting to the lakehouse via the One Lake catalog, then build a data model and prepare for Copilot-assisted data exploration in reports.
Prompting Microsoft Copilot to explain and document data19:24
Open a new report and prompt Copilot to describe the car rental dataset’s tables, columns, and business meanings, guided by the semantic model.
Using Copilot to identify and define Key Performance Indicators20:53
Discover how Copilot identifies KPIs for a car rental business, distinguishes data-driven versus derivable metrics, and defines formulas for fleet utilization and revenue per available car.
Using Microsoft Copilot to detect relationships in the data model16:01
Use Copilot to detect relationships in your data model and explain natural keys. Build a star schema and manage role playing dimensions with active and inactive relationships.

Asking Copilot business questions about the data5:58
Leverage Copilot to generate meaningful business questions about rental performance, fleet efficiency, branch performance, and revenue insights to guide dashboard measures.
Creating Basic DAX measures in Power BI using Copilot21:47
Copilot demonstrates creating basic DAX measures in Power BI, compares distinct count and count rows, and shows workflows from the DAX query view to desktop with live-query limits.
Creating Advanced DAX measures in Power BI using Copilot8:17
Leverage Copilot to create advanced DAX measures in Power BI for ai-powered analytics, building base metrics into fleet utilization, revenue per car, and branch occupancy while managing filter context and dates with datediff.

Prompt Copilot to create an executive overview dashboard page16:25
Use Copilot to create an executive overview dashboard for a car rental company, outlining pages and visuals like total revenue, reservations, and a monthly date slicer.
Prompt Copilot to create a branch performance dashboard page11:02
Explore using Copilot to build a branch performance dashboard page in Power BI, with selectors, slicers, and reusable metrics like reservations and revenue, plus tips for clear labeling and maintainability.
Prompt Copilot to create a fleet insights dashboard page14:29
Prompt Copilot to create a fleet insights dashboard page with meaningful visual titles, car counts by category, and metrics like average reservation duration by category and average car age.
Prompt Copilot to create a customer insights dashboard page13:49
Build a customer insights dashboard with Copilot in Microsoft Fabric and Power BI by creating a customer age metric, age bins, and visuals for reservations and revenue.
Trying to create a report design template7:52
Explore creating a dashboard theme with Copilot in Power BI and Fabric, generating a car rental light theme JSON, testing import, troubleshooting syntax errors, and learning AI limitations in design.

Conversational Analytics: Asking Copilot simple questions about our data8:46
Explore talk to data features that let you ask copilot simple questions. Receive time intelligence insights, year over year comparisons, and narratives revealing bookings and revenue by car model.
Conversational Analytics: Time Intelligence11:56
Explore time intelligence in Power BI with Copilot, analyzing monthly trends, peak and off-peak seasons, year-over-year revenue, and lead times using DAX measures and calendar concepts.
Conversational Analytics: Narratives and Summaries6:26
Use Copilot to generate narratives that summarize dashboard insights with data references, highlight monthly reservations and revenue trends, and link visuals for executive storytelling.

Data understanding and data exploration in Fabric Notebooks with Copilot23:19
Explore data understanding and exploration in Fabric notebooks with Copilot, loading lakehouse data, performing data quality checks and profiling, and visualizing key measures for data science workflows.
Advanced Analytics in Fabric Notebooks with Copilot16:25
Explore using Copilot in Fabric notebooks to perform advanced analytics, from linear regression and clustering to outlier detection and fraud analysis, with data preparation, feature engineering, and business interpretations.

Download Source Data and Dashboard0:01
Introduction to the Model Context Protocol2:09
Model Context Protocol vs. Application Programming Interfaces1:30
What the Power BI MCP Server can actually see1:28
Setting up Power BI MCP Server in Visual Studio Code9:28
Setting Up Power BI MCP Server in Claude Desktop5:17
Here is the configuration part that you have to add, remember to replace the folder:
"mcpServers": {
"powerbi-modeling-mcp": {
"command": "C:\\Users\\fschw\\.vscode\\extensions\\analysis-services.powerbi-modeling-mcp-0.4.0-win32-x64\\server\\powerbi-modeling-mcp.exe",
"args": [
"--start"
],
"env": {}
}
},
Exploring the Semantic Model and create a technical documentation using AI17:47
Here are the prompts used in the video:
First prompt:
List the connected data sources.
I am Power BI Report Developer and I have been handed over this undocumented report. I need the following:
- Analyse the data model
- Document the data model, entiites and attributes (provide description and definition incl. sample data for each)
- draw an er model for the data model
Write everything into a dedicated markdown file.
Other Prompts:
Do you think theres anything missing about the data model that would be relevant for a Power BI Report Developer? From a technical documentation point of view.
Analyse and add Power Query transformation lineage information
Create a business documentation of your Semantic Model using AI10:37
Here are the prompts used in this video:
Now I need a business documentation of FastBite. Please analyse the business model and give me a business description (non-technical) for each business entitiy. And create a comprehensive glossary. Target groups should be business, business analysts and management.
please add DAX formulas to the KPIs in Section 4.
Transformthe glossary into a table.
Do you think there's anything missing in the business documentation?
Download the Report Documentation Agent0:02
Using the report documentation agent to generate a full report documentation6:17

Data Modeling with Copilot/Claude: Develop your Semantic Model16:57
Here are the prompts:
List the power bi desktop instances
connect to the instance with the easy car dashboard
list the connected data sources
which tables do we have in our semantic model?
are there relationships defined or documented?
Perform an analysis of the tables in the semantic model and propse relationships. Can we model a star schema from the data?
Yes apply the changes to the semantic model. Create a markdown file where you document each step and modeling decision. It sohuld be a documetation of your work and of the data model. Keep this in mind for all future steps as well.
create a documentation for the easycar report
Developing DAX Measures using AI25:20
Here are the prompts:
First longer prompt:
Analyse the business and technical description for EasyCar. You are now something like a requirements engineer, whos job is to formulate reporting requirements from:
- executive management
- finance
- fleet ops
For each of these target groups, generate requirements and relevant business questions.
Write the output into an md file.

Other Prompts:
For each of the identified business questions, derive KPIs. Create a new KPI markdown document that maps each questions to a KPI. No DAX measures yet.
Which dashboards would we need? I do not think that a one fits all approach is suitable here.
write this in an md
For all the metrics you have identified in kpi catalog, perform an assessment, wheter these can be answered with the present data or if addiiotnal data is needed
For the KPI required for Dashboard 1 - executive cockpit - generate the DAX statements for each KPI
Can you implement and connect the missing date dimension into the semantic model?
implement the dax measures for the executive overview dashboard in the semantic
Exploring, Explaining & Understanding DAX Measures using Copilot/Claude4:28
Create a new markdown document for documenting this metric in detail. I want to know:
- What it does (business meaning)
- How it works
- What are potential issues
- visual dependency/hierarchy graph of all involved metrics
Let Copilot prepare Dashboard Mockups & Dashboard Development Instructions6:29
For the executive cockpit dashboard please create:
- page structure with visualisations and required KPI
- Mockup as SVG with realistic sample values
- For each visual: The Configuration incl. attributes, measures and specific settings. These should be a click instruction for the developers.
AI-supported Data Profiling5:21

Requirements

Curiosity to explore data and try out AI-powered workflows with Copilot.
No coding experience required — Copilot will help you write SQL, DAX, and PySpark.
An open mindset to experiment, prompt creatively, and let AI accelerate your work.
Access to an organizational Azure subscription is required if you want to work hands-on

Description

Imagine building dashboards, analyzing trends, writing DAX, documenting semantic models, and exploring your data simply by asking questions. No endless clicking, no staring at a blank formula editor, and no spending hours searching documentation. Just AI-powered analytics.

Artificial Intelligence is transforming the way we build, analyze, and consume data solutions. In this course, you will learn how to use Microsoft Copilot, Claude AI, the Power BI MCP Server, and Microsoft Fabric to dramatically accelerate analytics, reporting, semantic modeling, and business intelligence development.

You will discover how AI can help you generate DAX measures, SQL queries, and PySpark code; analyze and document semantic models; build reports and dashboards; explore data using natural language; and interact directly with Power BI and Fabric artifacts through AI features. You'll see how AI can turn plain English into working code, explain complex calculations, review models and reports, and help uncover insights hidden within your data.

Through practical, hands-on examples and end-to-end scenarios, you will learn how AI can support every stage of the analytics lifecycle. From data preparation and modeling to reporting, documentation, and insight generation.

Whether you are a Power BI Developer, Fabric Engineer, Data Analyst, BI Consultant, or Analytics Professional, this course will show you how to integrate AI into your daily workflow and build analytics solutions faster, smarter, and with greater confidence.

Who this course is for:

Power BI users who want to work faster, explore data conversationally, and automate tasks with Copilot.
BI analysts and data professionals looking to add AI-assisted analytics, DAX generation, and Fabric skills to their toolkit.
Beginners who want an easy, friendly introduction to Power BI and Fabric using natural language and Copilot guidance.
Data engineers who want to use Copilot for SQL, PySpark, data cleaning, and exploration inside Fabric Notebooks.
Students and career changers aiming to break into data analytics with modern, in-demand AI skills.
Anyone curious about AI in Microsoft’s data ecosystem and how Copilot transforms the way we analyze, model, and visualize data.

Microsoft Copilot, Fabric & Power BI: AI-Powered Analytics

What you'll learn

Explore related topics

Course content

Introduction2 lectures • 10min

Module 2: Setting up Microsoft Fabric, Power BI and Copilot4 lectures • 20min

Module 3: Understand the data by prompting Copilot in Microsoft Fabric & PowerBI7 lectures • 1hr 9min

Module 4: Defining measures, metrics and KPI in DAX using Microsoft Copilot3 lectures • 36min

Module 5: Create a complete dashboard in Microsoft Power BI using Copilot5 lectures • 1hr 4min

Module 6: Conversational Analytics with Copilot: Talk to your Fabric data!3 lectures • 27min

Module 7: Copilot for Data Science & Advanced Analytics in Fabric Notebooks2 lectures • 40min

Power BI Modeling Model Context Protocol (MCP) Server: Semantic Model Analysis10 lectures • 55min

Semantic Model & DAX Development using AI & LLM (GitHub Copilot)5 lectures • 59min

Module 8: Conclusion1 lecture • 5min

Requirements

Description

Who this course is for: