HANDS-ON GENERATIVE BI TRAINING
Get Your Power BI Models Ready for AI
A hands-on workshop that teaches Power BI teams how to turn existing semantic models into a more reliable foundation for Generative BI.
THE CHALLENGE
Accurate Dashboards Don’t Guarantee Accurate AI Answers
Power BI teams have spent years building semantic models that deliver accurate reports and dashboards.
But AI changes the interaction model.
Instead of clicking visuals and filters, business users ask questions in natural language..
AI has to interpret what the user means, choose the right measures, apply the correct date roles, understand model grain, and follow the right analytical rules.
A technically correct semantic model can still produce inaccurate or misleading AI responses.
Why?
Because much of the business knowledge and historical context Power BI professionals carry in their heads isn’t automatically available to the AI.
This workshop teaches your team how to close that gap.
WORKSHOP OVERVIEW
Prepare Your Existing Semantic Models for Generative BI
This full-day, hands-on workshop takes Power BI professionals through a practical process for improving semantic models for AI.
Using a realistic case study, attendees learn how to:
Evaluate AI readiness
Enrich semantic models with clearer business semantics
Simplify the AI data schema
Add business context and analytical rules
Configure verified answers
Test Copilot with realistic business questions
Validate, diagnose, refine, and retest
The course focuses on leveraging the Power BI assets your organization already has.
Your existing semantic models are the starting point.
WHAT YOUR TEAM WILL LEARN
Build Models That Give AI Needed Context
Recognize AI-Readiness Issues
Understand why technically correct semantic models can still fail when AI interprets business questions.
Communicate Business Rules
Use AI Instructions to provide business context, metric definitions, date-role rules, analytical guidance, and guardrails.
Enrich Business Semantics
Improve business-friendly names, descriptions, measures, terminology, and date-role clarity.
Configure Verified Answers
Learn when Verified Answers are useful, how to create them, and how they differ from AI Instructions.
Simplify the AI Data Schema
Use Power BI’s Prep data for AI features to choose the tables, columns, and measures AI should work with.
Test and Refine with Copilot
Ask realistic stakeholder questions, develop expected answers, validate Copilot’s responses, diagnose problems, and refine the model.
HANDS-ON LEARNING
Don’t Just Learn the Concepts. Apply Them.
This workshop centers on a realistic Power BI case study.
Attendees work through the process themselves:
Evaluate → Enrich → Simplify → Instruct → Verify → Test → Refine
You’ll begin by establishing a baseline for AI readiness.
Then you’ll progressively improve the semantic model, configure Prep data for AI, add AI Instructions and Verified Answers, and test the results with realistic stakeholder questions.
The goal is not just to understand the concepts.
It’s to leave with a repeatable process your team can apply to your organization’s Power BI semantic models.
WORKSHOP TOPICS
What the Workshop Covers
What makes a Power BI semantic model AI-ready
Why technically correct models can still fail with AI
Enriching models with business-friendly semantics
Simplifying the AI data schema with Prep data for AI
Adding AI Instructions, business context, and analysis rules
Creating and using Verified Answers
Testing Copilot with realistic stakeholder questions
Validating, diagnosing, refining, and retesting AI responses
Understanding current limitations and next steps with Fabric IQ and Fabric Data Agents
WHO SHOULD ATTEND
Built for Power BI and Analytics Professionals
This workshop is designed for professionals who build, manage, analyze, or work with Power BI semantic models, including:
Business and data analysts
BI and analytics developers
Power BI developers
BI and analytics managers
Data scientists
AI engineers
Anyone interested in building real-world Generative BI solutions
PREREQUISITES
What Attendees Should Already Know
Attendees should be familiar with:
Power BI Desktop
Power BI semantic models
Star schema concepts
DAX
Experience developing and publishing Power BI semantic models is recommended.
No prior experience with Power BI Copilot, Prep data for AI, Microsoft Fabric IQ, or Fabric Data Agents is required
WHAT YOUR TEAM WILL TAKE AWAY
A Practical AI-Readiness Method
By the end of the workshop, attendees should be able to:
Explain what makes a semantic model AI-ready
Recognize why technically correct models can still fail with AI
Enrich models with clearer business semantics
Simplify the AI data schema
Configure AI Instructions
Define business context and analysis rules
Create and use Verified Answers
Test Copilot with realistic business questions
Validate AI responses against expected answers
Diagnose whether a problem comes from the model or the AI configuration
Refine and retest the model
Understand current limitations
See how this work connects to Fabric IQ and Fabric Data Agents
HANDS-ON ENVIRONMENT
Technical Requirements
Because this is a hands-on workshop, attendees need access to the required Power BI and Microsoft Fabric environment.
Attendees must have:
Access to a Power BI subscription (e.g., the free trial)
Access to a paid Microsoft Fabric capacity, F2 or higher
Access to a Fabric workspace with appropriate permissions
Fabric capacity located in a region that supports Copilot
Detailed setup instructions are provided before the workshop and must be completed in advance.
If your organization restricts software installation, Microsoft Fabric access, or Copilot capabilities, attendees should coordinate with IT or a Fabric administrator before the workshop.
THE BIGGER PICTURE
Power BI Is the Beginning, Not the End
Preparing semantic models for AI is one step in a broader Generative BI journey.
The same focus on explicit business meaning becomes increasingly important as organizations explore:
Semantic enrichment
Conversational analytics
Microsoft Fabric
Fabric IQ
Ontologies
Fabric Data Agents
The workshop ends by connecting the Power BI work your team has completed to those broader Microsoft Generative BI capabilities.