Building Reports and Core Visuals

You open Power BI Desktop, click a new report page, and stare at a white rectangle. Your manager wants one thing by Friday: a single page that answers a plain question, “How are sales doing this quarter, and where?” The data is already loaded. Now you have to turn columns of numbers into something a busy person can read in ten seconds.

That white rectangle is the report canvas. This chapter is about filling it well.

Why this matters

A report is a page, or a set of pages, that shows your data as charts, numbers, and tables. You build one by dropping visuals onto the canvas. A visual is any single chart, number, or table on the page. Microsoft calls them “visuals” or “visualizations.” They mean the same thing.

The real skill is not clicking buttons. It is choosing the right picture for the question. A number that matters gets a big bold figure. A comparison between regions gets bars. A trend across months gets a line. Pick the wrong one and your reader either misreads the data or gives up.

Clean report pages get decisions made in meetings. Cluttered ones get ignored. The rest of this chapter gives you the core visuals, when to use each, and how to arrange them so the page reads at a glance.

The report canvas and the three panes

Open Power BI Desktop and select Report view (the icon on the far left that looks like a bar chart). You get a blank canvas in the middle and a set of panes docked on the right. Three panes do most of the work.

The Data pane lists every table and field you loaded. A field is one column of data, such as Region, Order Date, or Sales Amount. Older versions and some menus still call this the Fields pane.
The Visualizations pane is where you pick a chart type and feed it data. It has three tabs: Build visual, Format visual (the paintbrush icon), and Analytics (the magnifying-glass icon).
The Filters pane limits what a single visual, a page, or the whole report shows.

Field wells

Select a visual on the canvas, and the Build visual tab shows a set of empty boxes. These are field wells (Microsoft also calls them buckets). You drag a field from the Data pane into a well to tell the visual what to show.

The wells change with the visual type. A column chart shows X-axis, Y-axis, Legend, and Small multiples. A matrix shows Rows, Columns, and Values. A map shows Location.

One helpful behavior: when you drop a number field into a Values or Y-axis well, Power BI adds it up for you. Drop Sales Amount onto a chart split by Region and each bar shows the total sales for that region. That automatic math is called aggregation, and by default it is a sum. You can change it to average, count, or minimum later.

Try It Build a chart that compares sales across regions. You need data with one category column (like Region) and one number column (like Sales Amount).

Click a blank part of the canvas. In the Visualizations pane, on the Build visual tab, click the Clustered column chart icon. An empty placeholder appears on the canvas.

In the Data pane, find your category field (for example, Region) and drag it into the X-axis well.

Drag your number field (for example, Sales Amount) into the Y-axis well. Power BI draws one column per region, sized by total sales.

Click the three dots (More options) in the top corner of the visual, choose Sort axis, and sort by Sales Amount, descending.

What did you notice? With two drags and one sort, a table of numbers became a ranking anyone can read in a second.

Core visuals and when to use each

You do not need every visual. A strong report usually uses five or six well. Here are the ones to reach for and the question each one answers.

Card, for a single number. When one figure is the headline (total sales, market share, number of open tickets), a card puts it in big type with nothing to distract from it. Drag one measure into the Values well and you are done. The current card visual became generally available in the November 2025 Power BI release and replaces the older single card and multi-card visuals, though the old ones still work in existing reports.

KPI, for progress toward a goal. A KPI (key performance indicator) visual shows a value against a target and signals whether you are ahead or behind. Use it for “sales versus quota this quarter” rather than a bare number.

Column or bar chart, for comparing categories. Columns stand up; bars lie down. Both compare a number across a set of items: sales by region, tickets by team, revenue by product line. Bars (horizontal) are the better choice when category names are long, because the labels have room to sit.

Line chart, for a trend over time. When your question involves months, weeks, or days, a line shows the shape of the change. Put the date on the X-axis and the number on the Y-axis. Lines are for time; do not use them to compare unrelated categories.

Table and matrix, for detail. A table is a plain grid of rows and columns, good when people need exact values they can scan or copy. A matrix is like a pivot table in Excel: it crosses one set of categories down the side against another across the top, adds subtotals, and lets readers expand and collapse groups (for example, Region down the side, Product Category across the top, Sales in the cells).

Map, for geography. When the answer is “where,” a map beats a list of place names. Power BI offers several map types, including the Filled map, which shades regions by value, and add-ins such as ArcGIS for Power BI by Esri and Azure Maps. Drop a geographic field (Country, State, City) into the Location well and a number into the size or value well.

Slicer, for letting readers filter. A slicer is an on-page control, like a dropdown or a list of buttons, that lets a reader narrow the whole page to one region, one date range, or one product. You will build interactive filtering properly in Chapter 7; for now, know that the slicer is the visual that gives readers that control.

Formatting: making a visual readable

A visual with the right data can still be hard to read. Formatting fixes that. Select the visual, then click the Format visual tab (the paintbrush icon) in the Visualizations pane. The options change with the visual type.

Focus on a few high-value settings:

Title. Every visual needs a clear title that states what it shows, such as “Sales by Region, Q2.” Power BI fills in a default title from the field names, and the default is often clumsy. Rewrite it in plain words.
Data labels. Turning these on prints the actual number on each bar or point. They help when exact values matter. Turn them off when they crowd the chart.
Colors. Use color to carry meaning, not decoration. One color for all bars is fine when you are just ranking. Reserve a second color to highlight the one bar that matters (this year versus last year, your team versus the rest).
Axis titles. For a simple chart the axis labels are usually enough, and an extra axis title just adds clutter. For anything ambiguous, add one.

A good rule: if a setting does not help the reader answer the question faster, leave it off.

Page layout and alignment

Where you place visuals is part of the message. Readers scan a page top-left to bottom-right, so put the headline numbers across the top and the supporting detail below.

A few tools keep the page tidy:

Add pages from the Insert tab (New page), or with the + icon on the page tabs at the bottom of the canvas. Use separate pages for separate audiences rather than cramming everything onto one.
Gridlines and Snap to grid, on the View tab, help visuals line up cleanly instead of sitting at slightly different heights.
Align tools let you select several visuals and line up their edges or space them evenly, so the page looks deliberate rather than dropped together.

Leave white space. A page with room to breathe reads faster than one packed edge to edge.

In business: a sales overview page

Here is how the pieces fit for the manager’s Friday question, “How are sales doing this quarter, and where?”

+---------------------------------------------------------------+
|  [Card]        [Card]          [KPI]        [Slicer: Quarter]  |
|  Total Sales   Avg Order       Sales vs.                       |
|  $2.4M         $312            Quota  ▲                        |
+---------------------------------------------------------------+
|  [Line chart]                    |  [Filled map]              |
|  Sales by Month                  |  Sales by State            |
|  (the trend across the quarter)  |  (where the money is)      |
+----------------------------------+----------------------------+
|  [Bar chart] Sales by Product Line    [Matrix] Region x Month |
+---------------------------------------------------------------+

The top row answers “how are we doing” in three seconds: two cards for the headline numbers and a KPI for progress against the goal. The slicer lets the reader switch quarters without touching the report. The middle row answers “where,” with the trend line on the left and the map on the right. The bottom row holds the detail for anyone who wants to dig: a bar chart ranking product lines and a matrix that crosses region against month.

Every visual has a plain title. Nothing on the page exists just to fill space.

Common pitfalls

Chart junk. Backgrounds, heavy gridlines, 3-D effects, and drop shadows add nothing and slow the reader down. Strip decoration until only the data and its labels remain.

Pie charts with too many slices. A pie can show two or three parts of a whole. Past that, the slices look alike and nobody can tell 14% from 17%. If you have more than a few categories, use a bar chart and let the lengths do the comparing.

Wrong chart for the data. A line chart across product categories implies a trend that is not there, because there is no order or time between “Shoes” and “Hats.” Lines are for time. Bars are for categories. Match the chart to the shape of the question.

No titles, or default titles. A visual labeled “Sum of Sales Amount by Region” makes the reader translate. “Sales by Region” says it plainly. A page of untitled charts forces everyone to guess, and guesses turn into wrong decisions.

Misleading axes. Starting a bar chart’s value axis above zero makes small differences look huge. If you cut the axis, you exaggerate the story. Start value axes at zero unless you have a clear, stated reason not to, and never let the format hide what the numbers actually say.

Practice

Recreate the sales overview layout above with your own data (or the sample data that ships with Power BI Desktop). Aim for two cards, one line chart, one bar chart, and a slicer, each with a plain title.
Take one bar chart and format it two ways: once with data labels on and a highlight color for the top bar, once stripped to the plainest version. Decide which reads faster and why.
Find one chart in your draft that uses the wrong visual for its question (a line across categories, or a pie with six slices) and fix it.
Write the single question each page of your report answers, in one sentence. If you cannot, the page is trying to do too much.

Next, Chapter 6 covers DAX, the formula language you use to build custom numbers (like profit margin or year-over-year growth) when your data does not already contain them. Chapter 7 covers interactivity: slicers, cross-filtering, and drillthrough that let readers explore the report themselves.

Part of the Power BI track. Developed with AI assistance and reviewed by a human editor. Power BI changes often; menu names, features, and prices may have shifted since the “checked on” dates in this chapter. Verify anything critical against Microsoft’s official documentation.