A pie chart is a circle divided into slices, where each slice represents a category’s share of one whole. The bigger the slice, the bigger that category’s share. It is one of the most recognizable ways to show proportions, which is exactly why it gets overused, and also why it gets criticized so often.
A pie chart only works when your values genuinely add up to one total. If your categories do not represent parts of a single whole, a pie chart will visually imply a relationship that is not actually there.
In the Visualizations pane, click the Pie Chart icon. This adds an empty pie chart visual to the report canvas.

In the Visualizations pane, drag the number you want to measure, such as sales or count, into the Values section. This decides how large each slice is.

Drag the field you want to split your data by, such as product category or region, into the Legend section. This is what turns one solid circle into separate colored slices.

Click the Format icon to open the Format pane. Look for Data Labels and switch it on. This prints the value or percentage directly on each slice.

By default, a pie chart only shows the category name on each slice, which leaves people guessing at the actual numbers. In the Format pane, under Data Labels, the Label contents dropdown lets you choose Category, Data value, Percent of total, or a combination such as Category and percent of total, which is usually the single most useful option since it shows both what the slice is and how big its share actually is, without needing to check a legend at all.
It is also worth turning the legend off entirely once your labels are on the slices themselves. Reading a label directly on its slice is faster than glancing back and forth between a slice and a separate legend, and for a small number of categories, the slices alone carry enough information on their own.
A pie chart and a donut chart plot the same kind of data with the same Values and Legend fields, and the choice between them mostly comes down to style. A donut chart has a hollow center that can hold a total or a short label, which a pie chart cannot do since it uses the full circle for its slices. If you want that extra label in the middle, use a donut chart. Otherwise a plain pie chart is simpler and just as effective.
Using a pie chart when the numbers do not add up to a whole. If your categories overlap or measure different things entirely, a pie chart implies a part to whole relationship that does not actually exist in your data.
That covers building, formatting, and choosing a pie chart in Power BI. To keep building out your report with other chart types, explore Prwatech’s Power BI training program, which includes placement assistance.