Skip to content
Tutorials Center
My learning

Topics

Lesson 3 of 56 ยท Welcome to Power BI

A Tour of Power BI Desktop

Find your way around Power BI Desktop: the ribbon, the Report, Table, Model and DAX query views, the main panes, and how .pbix and .pbip files work.

Power BI Desktop looks busy the first time you open it: a ribbon on top, icons on the left, three panes on the right and a big empty canvas. Almost everything has a clear job, though, and once you know the layout you will stop hunting for buttons. This lesson is a map of the window, not a click-by-click exercise. The next lessons put the pieces to work.

The window is described in words because it is a desktop application. Menu names and layouts change a little with each monthly release, so treat the names here as landmarks and expect small differences on your screen. The runnable examples are stand-ins: they let you see what the views and file formats are made of using Python.

The big picture

The Desktop window

Ribbon Tabs of commands across the top.
View switcher Icons on the left change the view.
Canvas The page where visuals live.
Panes Filters, Visualizations and Data on the right.
1 / 2

The ribbon

The ribbon is split into tabs. The ones you use most are:

  • Home: Get data, Transform data (opens Power Query), Refresh, Publish, plus quick access to common visuals, new measure and new quick measure.
  • Insert: add visuals, text boxes, shapes, images and buttons to the canvas.
  • Modeling: relationships, new measure, new column, new table, sort by column, data type and formatting, and security roles (Manage roles).
  • View: themes, page view and gridlines, show or hide panes, mobile layout, and the Performance analyzer, Selection pane and Bookmarks pane.
  • Optimize: pause visuals while you edit, and tools such as the Performance analyzer shortcut.
  • Help: documentation, samples and community links.

Some tabs appear only when you select something. Click a column in the Data pane and a Column tools tab appears; click a measure and you get Measure tools. Click a visual and a Format or Build tab can appear. If a command seems to be missing, select the object it applies to.

The four views

The icons on the left edge switch the view. You always work on the same file, just looking at different layers of it.

Report view is the canvas where you build pages. Each page is one tab at the bottom of the window. You drag fields onto the canvas to create visuals, and format them with the Visualizations pane.

Table view shows the data that was loaded into the model, one table at a time, as a grid. It is read-oriented: you check values, data types and formats, add calculated columns and choose how a column is sorted. If a chart looks wrong, Table view is often the first place to look.

Model view shows each table as a box with its columns, and lines for relationships between tables. Here you create and edit relationships, hide fields from report authors, organise measures into display folders and set properties such as format and default summarisation. Chapter 4 covers it in depth.

DAX query view is a code editor for DAX. You write a query that starts with EVALUATE, run it, and see the resulting table below. It also has helpers for writing measures and for updating them back into the model. In the same family, newer versions of Desktop add a TMDL view that shows the whole model as editable text. You will use DAX query view to test and debug measures later in the track.

DAX
EVALUATE
SUMMARIZECOLUMNS (
    'Product'[Category],
    "Revenue", SUMX ( 'Sales', 'Sales'[Quantity] * 'Sales'[Unit Price] )
)
ORDER BY 'Product'[Category]

The query above is read-only here. It asks the model for one row per product category with a Revenue column. The pandas stand-in below does the same thing with a group-by.

Python Stand-in: what a DAX query returns

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The three panes

On the right side of Report view are the panes you will touch all day.

Filters pane. It lists filters at three levels: on this visual, on this page and on all pages. A filter applied at a level affects everything inside that level. You will use it for rules such as "only completed orders".

Visualizations pane. The top part is the gallery of visual types (column chart, line chart, table, matrix, card, map and many more). Below it are field wells such as Axis, Legend and Values, where you drop fields. The Format tab lets you change titles, colours, fonts and labels of the selected visual.

Data pane. It lists tables, and under each table its columns and measures. Icons tell them apart: a plain field is a column, a calculator-style icon is a measure, and a sigma sign next to a column means it will be summed by default when you add it to a visual. Hidden fields do not appear here in Report view.

Panes can be collapsed with the small arrows to give the canvas more room. The View tab brings them back.

The stand-in below shows the three filter levels with pandas. A visual-level filter changes one chart only, a page-level filter changes every chart on the page.

Python Stand-in: filter levels in the Filters pane

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Saving: .pbix, templates and projects

Press Ctrl+S and Desktop saves a .pbix file. A .pbix is a single package that contains the report layout, the queries from Power Query, the model definition and, for Import tables, the data itself. That makes it easy to email or copy, and also makes it opaque to version control, because it is a binary container rather than readable text.

Two other formats are worth knowing:

  • .pbit (template) keeps the queries, model and report but no data. When someone opens it, Desktop asks for parameter values and loads fresh data.
  • Power BI Project (.pbip) saves the report and the semantic model as folders of text files instead of one binary. It is designed for source control such as Git and for team development. Depending on your version it may be an option you enable first in the Preview features settings.

A .pbix is a ZIP-style package. The stand-in below builds a toy package with a few typical part names, to show the idea. It is not a real report file, and the real DataModel part is a compressed binary, not text.

Python Stand-in: a .pbix is a package of parts

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Save early and often, and keep a copy before large experiments. Desktop also recovers unsaved work after a crash, but a saved file is the only thing you can rely on.

Settings worth knowing

Open File > Options and settings > Options once and look around. A few to notice early: the Preview features list (new capabilities appear here before they are on by default), Data load (type detection and background data), Power Query Editor (show or hide the formula bar) and Report settings. Under Data source settings you manage credentials, which the next chapter explains.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. You want to see the loaded rows of a table and check a column's data type. Which view fits best?

    Table view shows the loaded data as a grid and lets you check types and formats. Model view shows tables and relationships, not rows.
  2. 2. Where do you create and edit relationships between tables visually?

    Model view draws tables as boxes and relationships as lines between them, and lets you edit them.
  3. 3. What is the main difference between a .pbix and a .pbip?

    A .pbix is one package file. A Power BI Project saves readable text files in folders so changes can be tracked in source control.
  4. 4. Where does a filter that should apply to every visual on one page belong?

    The Filters pane has three levels. A page-level filter affects all visuals on that page only.

Summary

  • The window has a ribbon, a view switcher on the left, a canvas and three panes on the right: Filters, Visualizations and Data.
  • The four views are Report (design), Table (inspect data), Model (relationships) and DAX query (run DAX). Newer builds also add a TMDL text view.
  • Filters work at visual, page and all-pages level.
  • A .pbix is one package with layout, queries, model and data; a .pbit is a template without data; a .pbip project stores text files for source control.
  • Contextual ribbon tabs appear only when you select a column, measure or visual.

Next, put the tour to use and make a chart in Build Your First Report in 10 Minutes.