Supercharge Power BI Power BI Is Better When You Learn to Write DAX (Matt Allington)（Z-Library）

# Supercharge Power BI: Power BI Is Better When You Learn to Write DAX ## 【One-Line Pitch】 A hands-on, practice-driven guide that teaches business users how to write DAX (Data Analysis Expressions) in Power BI, transforming them from report consumers into report creators who can build flexible, accurate data models and measures. Ideal for Excel users and business analysts who want to move beyond drag-and-drop reporting into real data analysis. ## 【Book Arc】 - **Opening (~0%–10%)**: The author establishes the book's core philosophy—you cannot learn DAX by reading alone; you must practice. He introduces the three learning areas (M Language, DAX Language, and analysis) and sets up the AdventureWorks sample database, a fictional retail bicycle company with five years of sales data, which serves as the practice environment throughout the book. - **Early (~10%–23%)**: The focus is on data modeling fundamentals. Readers learn to load data from Access databases, understand the critical distinction between lookup tables (dimensions) and data tables (facts), and adopt the "Collie layout methodology"—placing lookup tables at the top and data tables at the bottom in Relationships view. The star schema is introduced as the recommended data model shape. - **Early (~23%–32%)**: The book transitions from data preparation to DAX basics. Readers learn how to create, edit, and comment on measures, with emphasis on proper workflow habits (like placing measures in the correct tables). The author introduces IntelliSense as a debugging tool and stresses the importance of reusing measures within other measures. - **Middle (~32%–48%)**: Core DAX functions are covered in depth: SUM(), COUNT(), COUNTROWS(), MIN(), MAX(), COUNTBLANK(), and DIVIDE(). The author explains subtle differences between similar functions (like COUNT() vs. COUNTROWS()) and introduces DISTINCTCOUNT() for unique value counting. This section also introduces the critical concept of filter context—how matrices and slicers filter data before measures are evaluated. - **Middle (~48% onward)**: The book begins exploring how visuals interact with each other through cross-filtering, setting up the foundation for understanding CALCULATE() (covered in a later chapter). The emphasis shifts from writing individual measures to understanding how the data model and filter context affect measure results across different visual configurations. ## 【Key Takeaways】 - **Practice is non-negotiable** (Opening): The author explicitly states you cannot learn from this book by reading alone—you must do the exercises in front of your computer. The "Here's How" worked examples followed by practice exercises create a learn-by-doing loop that commits knowledge to memory. - **Lookup tables vs. data tables is the fundamental modeling distinction** (Early): Data tables contain transactional information (like sales), while lookup tables contain descriptive attributes (customers, products, calendar). Understanding this distinction is essential because it determines how measures behave when filtered by different visuals. - **The star schema is the recommended data model shape** (Early): While other shapes like snowflake schemas work, the star schema—with lookup tables at the top and data tables at the bottom—is optimal for performance and user comprehension. The visual layout methodology (Collie layout) helps business users intuitively understand relationships. - **Measure placement and naming habits matter** (Early): Creating measures via the Modeling tab can accidentally place them in wrong tables. The recommended approach is to create measures in context and use descriptive, long names. These habits save significant time debugging later. - **Comments and IntelliSense are your debugging allies** (Early): DAX supports both single-line (//) and multi-line (/* */) comments, which are invaluable for pausing work on complex formulas. IntelliSense stops prompting when syntax is wrong—a signal to stop and check your formula. - **Reusing measures is a core DAX best practice** (Early): Once a measure like [Total Sales] exists, it can be referenced in other measures (e.g., [Total Tax] = [Total Sales] * 0.1). This creates maintainable, modular calculations and reduces errors. - **COUNT() vs. COUNTROWS() have subtle but critical differences** (Middle): COUNT() ignores rows with blank values in the specified column, while COUNTROWS() counts every row regardless of blanks. Choosing the wrong function can produce incorrect results in datasets with missing values. - **Filter context is the key to understanding measure behavior** (Middle): A matrix filters data before evaluating measures—this is the "initial filter context." Understanding why [Total Number of Products] varies by product category while [Total Number of Customers] stays constant is the gateway to mastering CALCULATE() and advanced DAX. ## 【Reading Tips】 - **Skim the setup chapters (0%–10%)** if you already have Power BI Desktop installed and understand basic data loading. Focus instead on the data modeling concepts in the Early sections, which are foundational for everything that follows. - **Deep-read the filter context discussion (~48%)**—this is the conceptual turning point of the book. Understanding why measures behave differently across matrix rows is essential before moving to CALCULATE() and more advanced topics. - **Do every practice exercise**—the author's entire methodology depends on hands-on practice. If you're short on time, prioritize exercises for SUM(), COUNTROWS(), and DISTINCTCOUNT(), as these appear most frequently in real-world reporting. - **Watch for the 32-bit/64-bit Access connection issue** (Early): If you encounter the "Microsoft.ACE.OLEDB.12.0 provider is not registered" error, the author provides a blog link for resolution. This is a common stumbling block that can derail your practice session. - **Take note of the naming conventions and measure placement recommendations**—these habits may seem pedantic now, but they prevent significant confusion when you build larger, more complex reports. ## 【Coverage Limits】 This guide covers the book's opening through the middle sections (approximately 48% of the content), focusing on data modeling fundamentals, basic DAX functions, and the introduction of filter context. The excerpts do not cover later chapters on CALCULATE(), time intelligence functions, or advanced DAX patterns—these are referenced but not detailed in the available material. ##