In September 2025, I had the opportunity to attend the European Microsoft Fabric Community Conference held in Vienna, Austria. It was an inspiring event that brought together professionals, MVPs, and enthusiasts from across Europe to explore the latest advancements in Microsoft Fabric, Power BI, and data analytics.

My special thanks go to Ella Murphy , the Marketing Manager of the event, for her outstanding organization and warm hospitality. I’m also delighted to share that the next European Fabric Conference will take place in Barcelona in 2026 — you can find more details here: https://www.sharepointeurope.com/european-microsoft-fabric-community-conference/

During the Vienna conference, I had the pleasure of reconnecting with Zoe Douglas, whom I first met at the inaugural European Fabric Conference in Stockholm (2024) and later at the Microsoft MVP Summit 2025. I invited Zoe to deliver an insightful session for my Meetup community, and it turned out to be a great success. You can watch the full recording of her session here:

Watch the session recording: https://www.youtube.com/@IlgarZarbaliyev

As we continue learning and sharing within the Microsoft Fabric and Power BI communities, I’d love to invite you to join my Meetup group and subscribe to my YouTube channel for future sessions, events, and tutorials:

Meetup: Azerbaijan Power BI & Microsoft Fabric Community: https://www.meetup.com/azerbaijanpowerbi-microsoftfabriccommunity/

YouTube: https://www.youtube.com/@IlgarZarbaliyev

Finally, you can access the resource file used in this article and accompanying session from the following link:

Download resources: September, 2025 Resource Files

Documenting a Power BI semantic model is crucial for ensuring clarity, maintaining consistency, and empowering users to understand the data they are analyzing. Yet, it's a task often relegated to the "I'll do it later" pile because it can be tedious and time-consuming.

What if you could generate a comprehensive, dynamic data dictionary directly inside your Power BI Desktop with just a few lines of DAX? Thanks to the new INFO.VIEW functions, you now can. This article will show you how to automatically document your tables, columns, measures, and relationships, turning your model into a self-documenting system.

Calculation groups can greatly minimize the need for creating duplicate measures by enabling you to define DAX expressions as calculation items, which can then be applied to the existing measures in your model. In this newsletter article, I used the AdventureWorks dataset. The Power BI Desktop file, both for practice and with solutions, can be downloaded from the following link.

https://onedrive.live.com/?id=357FB5C8090FE1B2%21s1e4f1311b59d4cad85bac9e6d2a9107e&cid=357fb5c8090fe...

👀 Have you ever written a DAX measure with CALCULATE() and thought — “this should work!” — only to see it break unexpectedly? Or maybe you tried CALCULATETABLE() and magically, everything worked.

Both functions seem to answer the same question:

“How do I modify the filter context?”

But they work very differently — and using the wrong one can lead to confusing totals, blank visuals, and filters behaving oddly.

Today, let’s demystify CALCULATE() and CALCULATETABLE(), when to use each, and why CALCULATETABLE() is often the safer bet in complex calculations.

Have you ever written a DAX measure with HASONEVALUE() and thought — “This should work!” — only to see it break unexpectedly?

Or maybe you tried ISINSCOPE() and magically, everything worked.

Both functions seem to answer the same question:

“Am I at the leaf level of a hierarchy or not?”

Field Parameters in Power BI are a powerful feature that enables users to dynamically select and toggle between different dimensions or measures within a report visualization. This feature enhances user interaction by offering a simplified way to explore data without requiring complex DAX calculations. In this guide, we’ll delve into what Field Parameters are, why they’re important, and how to effectively use them in your Power BI reports.

You can find the data source file, along with Power BI files containing both solutions and those without, at the following link:

Sources Adding Color Measure

Integrating Salesforce with Power BI can feel like unlocking a treasure chest—rich data, powerful visualizations, and strategic insights just waiting to be explored. However, to truly leverage this potential, effective data modeling is paramount. 

Too often, teams connect their data hastily, excited to dive into dashboards—only to find themselves facing sluggish performance, mismatched figures, or confusing visuals. The culprit? Poor data modeling.

In this post, we’ll walk through the best practices for modeling Salesforce data in Power BI with real-world context to help you avoid the pitfalls and build a rock-solid foundation for insight. Let’s get started!

Understand Your Salesforce Data Model (Before You Even Open Power BI!)

Imagine starting a puzzle without knowing what the final picture looks like. That’s what connecting Power BI to Salesforce without preparation feels like.

Before opening Power BI, take a step back. Grab a whiteboard or notebook and map out the Salesforce objects and fields you plan to use.

For example, if your business wants to analyze sales performance, your journey might begin with Opportunities, then branch into Accounts, Contacts, and Users. However, each of these objects has its own rules and relationships, like how an Opportunity is linked to an Account, or how a User owns that Opportunity. Consider:

• Relationships: How are these objects related in Salesforce? Understand parent-child relationships (e.g., Account and Contact) and lookup relationships. This knowledge is crucial for creating accurate relationships in your Power BI model.
• Field Types: Be aware of the data types of your Salesforce fields (e.g., text, number, date, picklist). This will influence how you can transform and analyze the data in Power BI.
• Data Volume: How much data are you pulling? Large datasets might require more strategic modeling to ensure performance.
• Business Questions: What questions are you trying to answer with this data? This will help you focus on the relevant objects and fields and design your model accordingly.

Understanding this up front will save you countless hours later.

 Import Only What You Need

It's tempting to bring in every Salesforce object and field, "just in case." Resist this urge! Importing unnecessary data bloats your Power BI model, slows down refresh times, and can make analysis more complex. Instead:

• Be Selective: Only import the objects and fields directly relevant to your reporting needs.
• Consider Views: Leverage Salesforce list views to pre-filter data and bring in only the essential records.

• Optimize Queries: While the native Salesforce connector in Power BI does not support direct SOQL (Salesforce Object Query Language) queries, optimized data retrieval can still be achieved through alternative methods. To precisely control data selection and filtering, you can:

  • Develop a custom connector that supports SOQL queries, or

  • Use Web.Contents in Power Query to make REST API calls to Salesforce and pass SOQL queries manually.

  These approaches allow fine-grained control over data extraction, reducing volume and improving performance

Less is more when it comes to modeling for performance and clarity.

 Define Clear Relationships in Power BI

Once, a client couldn’t figure out why their reports showed $0 in revenue for certain accounts. The issue? A missing relationship between the Opportunities and Accounts tables.

In Power BI, relationships are everything. Without them, data lives in silos, and your visuals won’t reflect reality.

To prevent that:

• Identify Primary and Foreign Keys: Understand which fields uniquely identify records in each object (primary keys) and which fields link records between objects (foreign keys).
• Create Relationships: In Power BI's Model view, drag the primary key field from one table to the corresponding foreign key field in another.
• Define Cardinality: Specify the type of relationship (e.g., One-to-Many, Many-to-One, One-to-One). For example, one Account can have multiple Contacts (One-to-Many).
• Set Cross-filter Direction: Determine how filtering should flow between related tables (Single or Both). Choose "Both" when you need to filter one table based on selections in the other and vice versa.

Optimize Data Types for Performance and Analysis

Power BI does a decent job of guessing field types, but it’s not infallible. I once had a report that wouldn’t let me create a date hierarchy. Turned out the “date” column was formatted as text!

When modeling, always double-check:

• Numeric Fields: Ensure fields used for calculations are set to appropriate numeric types (e.g., Whole Number, Decimal Number).
• Date Fields: Verify date fields are recognized as dates to enable time-based analysis.
• Text Fields: Be mindful of large text fields, as they can impact performance. Consider if you need the entire text or if summarizing is sufficient.

5. Create Calculated Columns and Measures Thoughtfully

A data engineer once told me, “Calculated columns are like sugar—use sparingly.” And he was right.

Power BI gives you powerful tools to create new logic—whether row-by-row using calculated columns or aggregating using measures with DAX.

Here’s how to use them wisely:

• Use Calculated Columns: Use these to add new columns based on existing data within a table (e.g., concatenating first and last names). Be mindful of creating too many calculated columns, as they are computed during data refresh and can impact performance.
• Use Measures: These are calculations performed on aggregated data and are generally more performant than calculated columns for aggregations (e.g., calculating the total value of closed opportunities). Write clear and efficient DAX for your measures.

Handle Salesforce Picklist Fields Effectively

Picklists seem simple until they’re not. A single-select picklist (like "Stage" in Opportunities) is easy to use in Power BI. However, multi-select picklists (e.g., a field for "Interested Products" that might contain “Product A; Product B; Product C”) can be challenging for filtering and aggregation. Here’s how to handle them:

• Single-Select Picklists: These can typically be used directly in your Power BI model.
• Multi-Select Picklists: These often store multiple values in a single field (e.g., "Value 1; Value 2; Value 3"). You'll likely need to split these values into separate rows or columns in Power BI for effective analysis. Consider using Power Query to split these delimited values.

Manage Large Salesforce Datasets Strategically

A global sales team importing 10 million records daily quickly learned: performance isn’t just about the report—it starts at the data model.

If you’re working with large volumes:

• Enable Incremental Refresh: Configure incremental refresh in Power BI to only update new or changed data, significantly reducing refresh times.
• Use Dataflows: Consider using Power BI dataflows to prepare and transform your Salesforce data in the Power BI service before loading it into your datasets. This can improve performance and reusability.
• Aggregate Wisely: Pre-aggregate data in Power BI where appropriate to reduce the number of rows processed during analysis.

Document Your Data Model

At some point, someone will ask: “What does this field mean?” Or “Why did we join these two tables like that?”

If it’s not documented, your team could be guessing—or worse, changing logic they don’t understand. Therefore,

• Add Table and Column Descriptions: Provide clear descriptions for your tables and columns to help others (and your future self) understand the data model.
• Use Visual Relationship Diagrams: Power BI's Model view provides a visual representation of your relationships. Keep this organized and easy to understand.
• Comment your DAX Logic: Add comments to your DAX measures and calculated columns to explain the logic.

Final Thoughts: The Power is in the Preparation

Salesforce and Power BI are a powerful duo—but only when connected with intention.

Modeling isn’t just about tables and joins—it’s about aligning your data with your business goals. Done right, it empowers you to spot trends faster, measure performance more clearly, and make smarter decisions every day.

So, slow down at the start. Ask the right questions. Map the relationships. And build a model that tells the full story—accurately, efficiently, and with clarity.

Because the difference between a dashboard that dazzles and one that confuses often comes down to one thing: the data model behind  it.

#engagements #powerbi #powerbidesktop #powerquery #vizualization #salesforce #MVP #superuser #sfdc

DAX (Data Analysis Expressions) is at the heart of advanced analytics in Power BI, but writing complex DAX formulas can be intimidating for many users. Fortunately, Power BI now offers powerful automation tools that make DAX more accessible than ever. In this post, I’ll explain how you can speed up your data analysis and reporting by leveraging one-click DAX features—even if you’re not a DAX expert.

Power BI continues to evolve, and the Copilot feature is a game-changer for anyone looking to build insightful reports quickly. In this post, I’ll walk you through how to leverage Power BI Copilot for rapid, automated report creation—even if you have minimal experience with DAX or complex data modeling.

Learn how to build Custom Column and Conditional Columns in Power BI; in this concise tutorial I walk you through the Power Query Editor step by step.

Power BI is often seen as a reporting tool. But what if we viewed the data model not just as a technical structure, but as the foundation of user experience (UX)? In this post, we’ll explore how treating the Power BI semantic layer as a product interface can lead to more intuitive, sustainable, and user-friendly analytics solutions.

A Loan Calculator in Power BI allows users to explore different loan scenarios by adjusting key inputs such as loan amount, interest rate, and loan term. By using What-If Parameters, users can instantly see how changes affect monthly payments and total repayment amounts, enabling dynamic decision-making and clear financial insights—all in an interactive report environment.

You can find the data source file, along with Power BI files containing both solutions and those without, at the following link:

Study Materials 

Have you ever wanted to extract raw data from your Power BI semantic models directly into Python?

You're not alone—many developers and data engineers working with Microsoft Fabric and Power BI run into fragmented documentation and scattered solutions. That’s exactly why I wrote this blog.

In this step-by-step guide, you'll learn how to:

• Connect to a Power BI semantic model using adodbapi

• Authenticate securely with Microsoft

• Extract full tables into your Python environment

Whether you're automating workflows, validating data, or building external dashboards—this blog brings everything into one place.

Problem Statement

In many financial dashboards and models, daily interest rates — especially Treasury Bill (T-Bill) rates — are critical data points.
Traditionally, this information is either:

• Manually downloaded,

• Updated periodically via spreadsheets, or

• Hardcoded into static tables.

👉 However, manually maintaining this data introduces:

• Risk of errors,

• Time overhead,

• Lack of real-time updates.

In a dynamic world, an automated solution is needed to fetch daily updated Treasury Bill rates directly inside Power BI — without needing to maintain tables manually.