Copilot in Power BI Reports

Quick Answer: Copilot in Power BI uses generative AI to create narrative visuals (text summaries of data), suggest report page layouts, and generate entire report pages from natural language prompts. It requires Fabric/Premium capacity and responds to filter context, making summaries dynamic.

Copilot Capabilities for Report Authors

1. Create a Narrative Visual

The narrative visual uses Copilot to generate a natural-language summary of the data:

Insert tab → Narrative visual (with Copilot icon)

How it works:

• Copilot analyzes the semantic model and current filter context
• Generates a text summary highlighting key insights
• Updates dynamically as slicers and filters change
• Authors can customize the prompt to focus on specific aspects

Example prompts:

• "Summarize the key trends in revenue by region"
• "Highlight the top and bottom performing products"
• "Explain the year-over-year changes in profitability"
• "What are the main drivers of sales growth?"

2. Create a New Report Page with Copilot

Copilot pane → "Create a report page that shows..."

Copilot will:

• Select appropriate visuals based on data types
• Arrange visuals on the page
• Apply filters and formatting
• Add titles and labels

Example prompts:

• "Create a page showing sales performance by region with trend over time"
• "Build an executive summary with key KPIs, revenue by product, and monthly trends"
• "Design a customer analysis page with segmentation and geographic distribution"

3. Suggest Content for a Report Page

When adding visuals to a page, Copilot can suggest:

• Which measures to display
• Which dimensions to use for slicing
• Appropriate visual types for the data
• Layout arrangements for readability

4. Summarize the Semantic Model

Copilot can explain the underlying data:

• Table descriptions and relationships
• Available measures and their business meaning
• Data quality observations
• Suggested analyses based on the model structure

Narrative Visual Configuration

Customizing the Narrative

After Copilot generates the initial narrative:

1. Edit the prompt to focus on specific topics
2. Add specific references to measures or dimensions
3. Set tone (formal, casual, technical)
4. Include/exclude specific data points
5. Pin important insights that should always appear

Dynamic Behavior

The narrative visual responds to filter context:

• Changes when slicers are adjusted
• Reflects cross-filter selections from other visuals
• Updates when page-level or report-level filters change
• Can be configured to focus on specific filter combinations

Requirements and Limitations

Requirements

Limitations

• Copilot may not perfectly interpret complex data relationships
• Narratives should be reviewed for accuracy before sharing
• Custom visuals are not suggested by Copilot
• Very large models may have slower Copilot response times

Best Practices for Copilot in Reports

1. Review generated content — Copilot may misinterpret data context
2. Be specific in prompts — "Show revenue by region for 2026" is better than "show sales"
3. Use as a starting point — refine Copilot output rather than using it as-is
4. Combine with manual design — use Copilot for initial pages, then customize
5. Test with different filter contexts — ensure narratives remain accurate when filtered

On the Exam

The PL-300 frequently tests:

• Knowing what Copilot can and cannot do in Power BI reports
• Creating narrative visuals and understanding their dynamic behavior
• Using Copilot to suggest or create report pages
• Understanding the licensing requirements for Copilot features
• Recognizing when Copilot features are the appropriate solution