Power BI Interview Questions and DAX Scenarios (2026 Master Guide)

Q1. What are the core components of Microsoft Power BI architecture?

Foundation • 0-2 Yrs

Power BI consists of: Power BI Desktop (local authoring tool for ETL, modeling, and visualization), Power BI Service (SaaS cloud platform for sharing, scheduling, and workspace collaboration), Power BI Mobile (apps for iOS/Android), Power BI Gateway (bridges on-premise data sources to cloud), and Power BI Report Server (on-premises reporting).

Q2. Explain the difference between calculated columns and DAX measures.

Foundation • 0-2 Yrs

Calculated columns are evaluated row-by-row during data refresh and stored in memory (RAM) within the VertiPaq engine, consuming storage. Measures are dynamic DAX calculations computed on-the-fly at query time based on user interactions, slicers, and filter contexts without consuming disk/RAM storage.

Q3. What is the difference between Star Schema and Snowflake Schema?

Foundation • 0-2 Yrs

Star Schema features a central Fact table connected directly to de-normalized Dimension tables, which optimizes VertiPaq compression and query speed. Snowflake Schema normalizes dimensions into sub-dimensions (e.g., Product -> Subcategory -> Category), reducing data redundancy but increasing relationship joins and degrading analytical query speed.

Q4. What is Power Query and what language does it use under the hood?

Foundation • 0-2 Yrs

Power Query is the data transformation and mashup ETL engine inside Power BI and Excel. It uses the 'M' functional language (case-sensitive) to perform operations like unpivoting, merging, appending, datatype casting, and custom column creation before loading into the data model.

Q5. Explain the difference between Merge Queries and Append Queries in Power Query.

Foundation • 0-2 Yrs

Merge Queries joins two tables horizontally based on matching key columns (similar to SQL JOINs: Left Outer, Inner, Full Outer, etc.). Append Queries combines two or more tables vertically by stacking rows on top of each other (similar to SQL UNION ALL), provided columns match.

Q6. What are the different connectivity modes in Power BI (Import vs DirectQuery vs Live Connection)?

Foundation • 0-2 Yrs

Import Mode loads data directly into Power BI's in-memory VertiPaq engine (highest performance, supports all DAX, 1GB dataset limit per Pro workspace). DirectQuery keeps data at the source and sends native queries per visual interaction (real-time data, slower visuals, limited DAX). Live Connection connects exclusively to SSAS or Power BI datasets without data modeling.

Q7. What is Row-Level Security (RLS) and how is it implemented?

Foundation • 0-2 Yrs

RLS restricts data access for specific users based on roles. In Power BI Desktop, roles are defined with DAX filter rules (e.g., [Region] = USERNAME() or USERPRINCIPALNAME()). Once published to Power BI Service, users or security groups are assigned to those roles.

Q8. Explain the difference between USERNAME() and USERPRINCIPALNAME() in Power BI.

Foundation • 0-2 Yrs

In Power BI Desktop, both return domain\username. In Power BI Service, USERNAME() returns the user's login name, while USERPRINCIPALNAME() strictly returns the user's email address (e.g., user@company.com), making it standard for dynamic RLS implementations.

Q9. What is the VertiPaq engine and how does it compress data?

Foundation • 0-2 Yrs

VertiPaq is the in-memory columnar database engine that powers Power BI and Analysis Services. It achieves 10x-20x compression through Columnar Storage, Value Encoding, Dictionary Encoding (Hash Encoding), and Run-Length Encoding (RLE) on sorted high-cardinality columns.

Q10. What are the primary building blocks of a Power BI report?

Foundation • 0-2 Yrs

Visualizations (charts, matrices, cards), Datasets (tables, relationships, measures), Reports (multi-page collections of visuals), Dashboards (single-page executive summaries in Power BI Service composed of pinned tiles), and Tiles (individual visual snapshots).

Q11. What is the purpose of the Performance Analyzer in Power BI Desktop?

Foundation • 0-2 Yrs

Performance Analyzer logs and measures the duration (in milliseconds) required to render each report visual, broken down into: DAX Query time, Visual Display rendering time, and Other (internal queueing and data preparation), allowing analysts to pinpoint slow DAX measures.

Q12. Explain the difference between implicit and explicit measures in Power BI.

Foundation • 0-2 Yrs

Implicit measures are automatically created by Power BI when you drag a numeric field into a visual and select a default aggregation (Sum, Average, Count). Explicit measures are user-written DAX formulas using the measure formula bar, which are reusable, maintainable, and support advanced logic.

Q13. What is a Date Table and why is it mandatory for Time Intelligence in DAX?

Foundation • 0-2 Yrs

A Date Table (Calendar table) contains contiguous rows for every date across the dataset range without gaps or duplicates, marked as a 'Date Table'. It is mandatory for Time Intelligence functions (YTD, SAMEPERIODLASTYEAR, DATEADD) to accurately calculate calendar offsets without skipping dates.

Q14. What are Quick Measures and when should they be used?

Foundation • 0-2 Yrs

Quick Measures are pre-built templates provided by Power BI to generate complex DAX formulas (like running totals, percentage differences, year-over-year change) without writing DAX from scratch. They serve as excellent learning scaffolding for junior analysts.

Q15. What is the difference between ALL() and ALLEXCEPT() functions in DAX?

Foundation • 0-2 Yrs

ALL(Table[Column]) removes all filter context from specified columns or the entire table, returning all rows regardless of slicers. ALLEXCEPT(Table, Table[Column]) removes filters from all columns in the table EXCEPT the specified columns.

Q16. Explain Evaluation Context in DAX: Row Context vs Filter Context.

Experienced • 3-5+ Yrs

Row Context exists during row-by-row iteration (in calculated columns or iterator functions like SUMX, FILTER); it knows the 'current row' but does not filter the table. Filter Context is the set of all active filters applied to the data model via slicers, visual coordinates, page filters, and CALCULATE. DAX measures evaluate strictly inside Filter Context.

Q17. What is Context Transition and how does CALCULATE trigger it?

Experienced • 3-5+ Yrs

Context Transition occurs when a Row Context is converted into an equivalent Filter Context. Calling CALCULATE() (or invoking a measure, which has an implicit CALCULATE wrapped around it) inside an iterator (like SUMX or a calculated column) transforms the current row values into active filters on all columns of that table.

Q18. Explain the internal execution of CALCULATE() in DAX.

Experienced • 3-5+ Yrs

CALCULATE evaluates its expression in a modified filter context: 1. It makes a copy of the original filter context. 2. It performs context transition if executed inside a row context. 3. It evaluates filter arguments. 4. It resolves filter conflicts (overwriting or merging with original filters). 5. It evaluates the base expression in the newly constructed filter context.

Q19. What is the difference between USERELATIONSHIP() and CROSSFILTER() in DAX?

Experienced • 3-5+ Yrs

USERELATIONSHIP() activates an inactive relationship (dotted line) for the duration of a specific CALCULATE calculation (e.g., using ShipDate instead of OrderDate). CROSSFILTER() dynamically alters the cross-filtering direction (Single, Both, or None) of an existing active relationship.

Q20. How does KEEPFILTERS() modify CALCULATE filter behavior?

Experienced • 3-5+ Yrs

By default, filter arguments inside CALCULATE overwrite existing filters on the same column. Wrapping the filter in KEEPFILTERS() instructs CALCULATE to intersect the new filter with existing filters rather than overwriting them, preserving ambient slicer selections.

Q21. Explain the difference between SUM() and SUMX() in DAX.

Experienced • 3-5+ Yrs

SUM() is an aggregator that works on a single column in the current filter context. SUMX() is an iterator that takes a table as its first argument, establishes a row context for each row of that table, evaluates an expression (e.g., Quantity * UnitPrice), and then sums the evaluated values across the rows.

Q22. How do you handle bidirectional relationships safely and avoid ambiguity?

Experienced • 3-5+ Yrs

Bidirectional relationships can introduce ambiguous filter paths, circular dependencies, and unexpected filter propagation. Best practices recommend keeping physical relationships single-direction and activating bidirectional filtering selectively using CROSSFILTER(..., Both) inside specific measures.

Q23. What is Query Folding in Power Query and why is it critical for performance?

Experienced • 3-5+ Yrs

Query Folding is the ability of Power Query to translate user transformation steps (filter, select, group by, joins) into a single native SQL query executed directly on the source database server. This leverages source database indexing and prevents loading millions of unneeded raw rows across the network.

Q24. Explain Incremental Refresh and how RangeStart and RangeEnd parameters work.

Experienced • 3-5+ Yrs

Incremental Refresh partitions large tables into historical and recent partitions. RangeStart and RangeEnd (DateTime parameters) filter the source data. Only data in the recent refresh window (e.g., last 10 days) is refreshed, while historical partitions remain static, slashing refresh times from hours to seconds.

Q25. What is the difference between RELATED() and RELATEDTABLE() in DAX?

Experienced • 3-5+ Yrs

RELATED() navigates from the Many-side to the One-side along an active relationship to fetch a single scalar value from a dimension column. RELATEDTABLE() navigates from the One-side to the Many-side, returning an entire sub-table of matching rows (often passed into iterators like COUNTROWS).

Q26. How do you optimize slow DAX measures using DAX Studio?

Experienced • 3-5+ Yrs

In DAX Studio, connect to the Power BI model, run Server Timings, and inspect Formula Engine (FE) vs Storage Engine (SE) duration. If FE is high (>20%), the measure is doing excessive single-threaded iterator computations; refactor to push logic down to VertiPaq's multi-threaded Storage Engine (xmSQL).

Q27. What are Composite Models and Aggregations in Power BI?

Experienced • 3-5+ Yrs

Composite Models allow combining DirectQuery and Import mode tables in the same model. Aggregations pre-calculate and import summarized data (e.g., Sales by Month and Region) while maintaining DirectQuery for granular transaction-level details; queries seamlessly hit in-memory aggregates for speed.

Q28. Explain the TREATAS() function and how it achieves virtual relationships.

Experienced • 3-5+ Yrs

TREATAS applies the result of a table expression as filters to columns of an unrelated table without defining a physical relationship in the model, enabling virtual joins and high-performance cross-filtering between disconnected tables.

Q29. How do you implement Dynamic M Query Parameters for DirectQuery sources?

Experienced • 3-5+ Yrs

Dynamic M Query Parameters allow report consumers to pass slicer or filter values directly into M parameters inside a native SQL query, enabling the database to filter partitions dynamically before returning data to Power BI.

Q30. Explain the difference between EARLIER() and DAX Variables (VAR/RETURN).

Experienced • 3-5+ Yrs

EARLIER() accesses an outer row context during nested iterations, which is verbose and hard to maintain. Modern DAX replaces EARLIER by declaring variables (VAR) in the desired row context; variables are evaluated once and hold their value as constants throughout the RETURN statement.

Q31. Write a DAX measure to calculate Year-to-Date (YTD) Sales using TOTALYTD.

Live Coding Challenge

We use TOTALYTD referencing the base sales measure and the contiguous calendar date column.

Total YTD Sales = 
TOTALYTD(
    [Total Sales],
    'DimDate'[DateKey]
)

Q32. Write a DAX measure to calculate Year-over-Year (YoY) Sales Growth Percentage.

Live Coding Challenge

We calculate previous year sales using SAMEPERIODLASTYEAR and compute percentage change using DIVIDE.

YoY Sales Growth % = 
VAR CurrentSales = [Total Sales]
VAR PriorYearSales = 
    CALCULATE(
        [Total Sales],
        SAMEPERIODLASTYEAR('DimDate'[DateKey])
    )
RETURN
DIVIDE(CurrentSales - PriorYearSales, PriorYearSales, 0)

Q33. Write a DAX measure to find the Moving 3-Month Average Sales.

Live Coding Challenge

We use DATESINPERIOD with a -3 month interval to average sales dynamically.

Moving 3M Avg Sales = 
VAR LastSelectedDate = MAX('DimDate'[DateKey])
VAR Period3M = 
    DATESINPERIOD(
        'DimDate'[DateKey],
        LastSelectedDate,
        -3,
        MONTH
    )
RETURN
CALCULATE(
    AVERAGEX('DimDate', [Total Sales]),
    Period3M
)

Q34. Write a DAX measure to calculate Sales for Top 5 Customers by revenue.

Live Coding Challenge

We use CALCULATE with TOPN filtering customers by their sales rank.

Top 5 Customer Sales = 
CALCULATE(
    [Total Sales],
    KEEPFILTERS(
        TOPN(
            5,
            ALL('DimCustomer'[CustomerName]),
            [Total Sales],
            DESC
        )
    )
)

Q35. Write a DAX formula to ignore visual slicer filters on the Product Category.

Live Coding Challenge

We wrap the base calculation inside CALCULATE using ALL on the category column.

All Category Sales = 
CALCULATE(
    [Total Sales],
    ALL('DimProduct'[ProductCategory])
)

Q36. Write a DAX measure to calculate Cumulative Running Total Sales over time.

Live Coding Challenge

We filter dates up to the maximum date visible in the current filter context.

Running Total Sales = 
VAR CurrentDate = MAX('DimDate'[DateKey])
RETURN
CALCULATE(
    [Total Sales],
    FILTER(
        ALL('DimDate'[DateKey]),
        'DimDate'[DateKey] <= CurrentDate
    )
)

Q37. Implement dynamic currency conversion in DAX using a disconnected conversion table.

Live Coding Challenge

We lookup the selected currency rate from a disconnected slicer table and multiply.

Converted Sales = 
VAR SelectedRate = 
    SELECTEDVALUE('DimExchangeRates'[RateToUSD], 1.0)
RETURN
[Total Sales USD] * SelectedRate

Q38. Write a DAX measure to count customers who purchased in BOTH 2025 and 2026.

Live Coding Challenge

We use INTERSECT on customer lists calculated across both calendar years.

Repeat Customers 2025-2026 = 
VAR Cust2025 = 
    CALCULATETABLE(
        VALUES('FactSales'[CustomerID]),
        'DimDate'[Year] = 2025
    )
VAR Cust2026 = 
    CALCULATETABLE(
        VALUES('FactSales'[CustomerID]),
        'DimDate'[Year] = 2026
    )
RETURN
COUNTROWS(INTERSECT(Cust2025, Cust2026))

Q39. Write an M query snippet in Power Query to unpivot monthly columns into attributes.

Live Coding Challenge

We invoke Table.UnpivotOtherColumns to normalize wide monthly financial tables into tabular form.

let
    Source = Excel.Workbook(File.Contents("C:\Data\Finance.xlsx"), null, true),
    Sheet1 = Source{[Item="MonthlyData",Kind="Sheet"]}[Data],
    PromotedHeaders = Table.PromoteHeaders(Sheet1, [PromoteAllScalars=true]),
    UnpivotedCols = Table.UnpivotOtherColumns(PromotedHeaders, {"AccountID", "Region"}, "Month", "Amount")
in
    UnpivotedCols

Q40. Write a DAX measure for Dynamic Row-Level Security verifying user email against manager hierarchy.

Live Coding Challenge

We check if user email matches either the employee or their reporting manager.

-- Placed inside Manage Roles DAX Filter Expression
[EmployeeEmail] = USERPRINCIPALNAME() || [ManagerEmail] = USERPRINCIPALNAME()