What is Power BI?
Power BI is Microsoft's business intelligence platform: you connect to data, clean it, model it, write calculations in DAX and build interactive reports that others can explore. Its main parts are Power BI Desktop (the free Windows app where reports are built), the Power BI Service (app.powerbi.com, where they are published, refreshed and shared) and the mobile apps. Power BI is now also part of Microsoft Fabric.
What are the building blocks of Power BI?
Semantic models, visuals, reports, dashboards and tiles. A semantic model (called a dataset before 2023) holds the tables, relationships and measures. A report is one or more pages of visuals built on one semantic model. A dashboard is a single page in the Service made of tiles pinned from reports. Workspaces and apps hold and share all of them.
What is the difference between Power BI Desktop and the Power BI Service?
Desktop is where you build: get data, Power Query, the model, DAX and the report pages, saved as a
.pbixfile. The Service is where the work lives once published: workspaces, apps, sharing, scheduled refresh, dashboards, subscriptions and security roles assigned to people. You can edit reports in the Service, but modeling is still done mostly in Desktop.What is the difference between a report and a dashboard?
A report is built on one semantic model, can have many pages, and is fully interactive: slicers, cross filtering, drill down. A dashboard exists only in the Service, is a single canvas of tiles pinned from one or more reports, and has no slicers. Dashboards are for a quick status view and for data alerts on card, KPI and gauge tiles; reports are for analysis.
Walk me through how raw data becomes a shared report.
Get data from the source, shape it in Power Query (types, filters, merges), load it, then build the model: a star schema with relationships and a date table. Write the measures in DAX, build the report pages, and publish to a workspace. In the Service, set credentials and a refresh schedule (through a gateway for on-premises sources), then share it through an app or with the people who need it.
What are Power Query and DAX, and when do you use each?
Power Query, written in the M language, shapes data before it loads: removing columns, fixing types, merging and appending tables. DAX calculates on the loaded model: measures such as totals, ratios, year to date and rankings that change with every filter. The usual rule is to clean and reshape in Power Query (or the source) and to aggregate in DAX.
What is the difference between Import mode and DirectQuery?
Import copies the data into the model's in-memory engine (VertiPaq): visuals are fast, every DAX and Power Query feature works, but the data is only as fresh as the last refresh and the model has a size limit (1 GB per model with a Pro license). DirectQuery stores no data and sends a query to the source each time a visual loads: the data is current, but visuals are slower, the source takes the load, and some transformations and DAX functions are limited. A composite model mixes both, and Fabric adds Direct Lake.
Which visual would you use for a trend, a comparison, a share of a total and a single number?
A line chart for a trend over time, a bar or column chart sorted by value for a comparison, a stacked bar or a treemap for a share of a total (a pie or donut only with a few slices), and a card or KPI visual for a single number. Use a table or matrix when people need exact values, and a map only when location matters to the question.
What is the difference between a slicer and the Filters pane?
A slicer is a visual on the canvas that the reader clicks, and it can be synced across pages. The Filters pane holds filters at three levels (one visual, one page, every page) plus drillthrough filters; it takes no space on the canvas, can be hidden or locked, and has advanced options such as Top N on a visual. Use slicers for the choices readers make often, and the Filters pane for the rest.
What Power BI licenses are there?
Free lets you build in Power BI Desktop and use your own My workspace, but not share. Pro lets you publish to shared workspaces and share with other Pro users. Premium Per User (PPU) adds Premium features per person, such as larger models, more refreshes and deployment pipelines. A capacity (Fabric F SKUs, which replace the older Premium P SKUs) runs content for an organization, and at F64 or above free users can view what is published to it.
What is the difference between a .pbix, a .pbit and a .pbip file?
A
.pbixis a report with its model and the imported data in one file. A.pbitis a template: the queries, model and report without the data, so whoever opens it connects to their own source. A.pbipis a Power BI project: a folder of text files (the model in TMDL, the report in PBIR) made for Git, code review and comparing versions.
Power BI Interview Questions for Freshers
What Power BI is, its parts, how a report gets from raw data to the Service, and the first terms every round checks.