Interview QuestionsJul 16, 20268 min read
Power BI Interview Questions: DAX, Modelling & Real Scenarios
Power BI interviews reward candidates who can move between three layers fluently: Power Query (M), the data model, and DAX. Interviewers deliberately design questions that require touching all three — surface-level DAX knowledge alone will not pass a serious BI-developer loop.
Data modelling fundamentals
- Star schema vs snowflake — why does Power BI prefer star?
- Difference between fact and dimension tables — how do you decide which is which?
- Explain single-direction vs bi-directional cross-filter relationships and when bi-directional causes problems.
- Role-playing dimensions — model a Date table used for OrderDate, ShipDate, and DeliveryDate.
- Composite models: when would you mix DirectQuery and Import in one model?
DAX questions that separate candidates
- Difference between CALCULATE and CALCULATETABLE — with an example.
- Row context vs filter context — walk through an example where they conflict.
- Write a YTD, MTD, and prior-year measure using DATESYTD, DATEADD, or SAMEPERIODLASTYEAR.
- How do you write a running total that respects the current filter context?
- USERELATIONSHIP — when do you use it and what breaks if you forget it?
- Explain VAR, RETURN — why does it make DAX both faster and more readable?
- Difference between SUM(column) and SUMX(table, expression).
Power Query (M) scenarios
- Unpivot a report where months are columns.
- Merge two queries — when do you use Merge vs Append?
- Handle changing source column names without breaking downstream steps.
- Explain query folding and how you'd verify it's happening.
- Parameterise a data source URL so the same report works across dev/UAT/prod.
Performance & optimisation
- A visual takes 40 seconds to render — walk me through your investigation.
- Why is a wide table with 50 columns worse than 5 narrower tables joined together?
- When would you use aggregations tables, and how do you set them up?
- Difference between Import, DirectQuery, and Live Connection — with a use case for each.
Deployment, governance, RLS
- How does Row-Level Security work in Power BI — static vs dynamic RLS?
- Difference between workspaces, apps, and deployment pipelines.
- How would you version-control a .pbix file that's binary?
- Refresh scheduling: gateway vs cloud, and how do you handle a failed refresh at 2 AM?
When you don't know the exact DAX syntax, describe the pattern and the filter-context change you want. Interviewers care that you know what to do; they'll accept 'I'd wrap this in CALCULATE with a filter that removes the year filter'.
Recommended next action
Take the next concrete step — it's free, takes under a minute, and gives you a real score to act on.
Practice BI analyst interviews