A pie chart shows how a total is divided, such as how spending is split across budget categories or which topics account for customer support requests. With a few clearly labeled slices, readers can see which categories contribute most and how much of the total each represents.
Making one starts with a simple table: the categories you want to compare and an amount for each. This guide shows you how to make a pie chart in Excel and other tools, using the same customer support example throughout. We’ll start with Deepnote’s chart-maker workflow, then cover spreadsheet steps, a short Python example, and PowerPoint.
Make a pie chart quickly with Deepnote
For a quick start, open Deepnote’s pie chart maker and drag in a CSV or Excel (.xlsx) file. You can also use the upload button to choose a file from your computer. If you already have a small table of prepared totals, you can paste the categories and values instead. Then describe the chart you want. You do not need to calculate the percentages first.
Imagine a support team reviewing 200 customer requests received in April. Each request has exactly one primary topic, so the categories below divide the full set without counting any request twice. These numbers are made up for the tutorial.
Copy the text below into the chart maker. It already includes the example numbers.
Create a 2-D pie chart from these fictional counts of customer support requests received in April. Each request belongs to exactly one primary topic.
Request topic Requests Account access 80 Billing 60 Product questions 40 Other 20 Use Requests to determine slice sizes. Label each slice with its topic and percentage of the total, using whole percentages. Title the chart “Customer support requests by topic, April”. Keep the categories in the supplied order and verify that the counts total 200.
Account access is the largest topic, but it accounts for less than half of all requests. Focusing on that topic alone would leave most requests unaddressed. Its 40% label also gives you a quick check that the chart matches the source data: 80 of the 200 requests concern account access.
You can follow up with “Make this a donut chart” or ask for different labels. If you have not settled on a chart type, start with the broader AI chart generator.
How to make a pie chart in Excel
Start with a category column and one column of amounts, such as request counts, sales, or spending. Use values that describe parts of the same total. In our example, that total is all 200 requests received in April.
The examples shown use the browser versions of Excel and PowerPoint. Instructions for the desktop versions are also included where the steps differ.
Make one in four steps
1. Arrange your data in two columns. Put the request topics in the first column and their counts in the next, with a heading for each. Use one row per topic, as in the example above. You can place the table anywhere on the sheet.
2. Select the categories and counts. Highlight both columns, including their headings and the four data rows. Leave out any grand total row: the total is the whole pie, not another category.
3. Insert the chart. Open Insert, choose the pie chart option, and select a flat, 2-D pie. Excel creates one slice for each topic. In desktop Excel, the menu is Insert > Insert Pie or Doughnut Chart > 2-D Pie.
4. Name the chart. Replace the default title with “Customer support requests by topic, April.” This tells readers what the slices represent and which period they cover. Next, add the labels that make each share readable.
If an unexpected slice takes up half the circle, check whether you included a total row. In this example, selecting the total of 200 alongside the four categories would count the same requests twice. Remove that row from the chart’s data range.
Add readable labels and percentages
In Excel for the web, select the chart and open its Format pane. Expand the series settings, turn on Data Labels, and choose percentage labels. Keep the legend visible when the labels show only percentages, so readers can identify each topic.
In desktop Excel, use Chart Design > Add Chart Element > Data Labels > More Data Label Options. Select Category Name and Percentage, and clear Value if you do not need the request counts displayed as well.
Show each topic’s name and percentage, either together beside the slice or through percentage labels and a clear legend. Readers should be able to identify each share without guessing from its color. Adding the request count to every label is unnecessary for this comparison.
- Use whole percentages here. All four shares are exact multiples of 10%. Add decimal places only when they help distinguish meaningful differences.
- Move crowded labels outside. Connecting lines can make it clear which label belongs to which slice. Enlarge the chart before shrinking the text.
- Remove a redundant legend. If every slice already names its category, readers do not need a second color key.
Keep the original request counts in your table. When you enable percentage labels, Excel calculates each category’s share of the total. For example, 80 account-access requests out of 200 appear as 40%.
Make the donut variant
A donut chart shows the same shares with an opening in the center. Excel calls it “Doughnut.”
- In Excel for the web, choose Doughnut from the chart-type options on the Chart tab. In the Format pane, expand the series settings to adjust Doughnut Hole Size.
- In desktop Excel, select Chart Design > Change Chart Type > Doughnut.
Use the donut when you want room for a total or short label in the center. Leaving that space empty is also fine. Keep the ring thick enough for the slices and their labels to remain easy to read.
Exploded and pie-of-pie charts: when to use them
An exploded pie chart separates one or more slices from the circle. In desktop Excel, to emphasize one slice, select the chart, then select that slice and drag it slightly outward. To separate every slice, open Format Data Series and adjust Pie Explosion.
A small separation can draw attention to the category being discussed. Usually, a contrasting color and a clear label are sufficient. Pulling every slice apart adds gaps without helping readers compare the shares.
A pie of pie chart breaks selected small slices out into a second pie. In desktop Excel, choose Chart Design > Change Chart Type > Pie > Pie of Pie. Use Format Data Series > Split Series By to control which values appear in the second plot.
The second pie shows the breakdown of a combined slice in the main pie. It is not another total to add to the first one. Label that relationship clearly. Our four-category example does not need this treatment; use it only when the breakdown of several small categories is important enough to justify the extra chart.
If your spreadsheet has one row per request
Real datasets often contain one row per request rather than one row per category. In that case, summarize the records before creating the pie chart.
Select the data and choose Insert > PivotTable. Put Request topic in Rows, then add it to Values and summarize it by Count. The result gives you one count per topic, which you can use as the source for the chart.
Check the records before charting. Decide how to handle blank topics, and use a request ID to investigate accidental duplicate records if each request should appear only once. Repeated topic names are expected; repeated copies of the same request may not be.
Create the pie from the PivotTable counts and leave the grand total out of the chart range.
If the pie chart is part of a recurring report
Keep the underlying records in an Excel table with fields such as request date, team, request ID, and primary topic. Build the PivotTable and chart from that source rather than copying category totals into a new sheet each month.
Add a filter or slicer for fields such as month or team. When you filter the PivotTable, the chart should represent the currently included records, so interpret each percentage against that filtered total rather than the all-time dataset.
When new records are added, refresh the PivotTable before using the chart in the next report. Check the filtered total as well as the percentages so a missing category, blank value, or incomplete refresh does not silently change the story.
How to make a pie chart in Google Sheets
Paste the request topics and counts into two adjacent columns in Google Sheets. In a desktop browser:
- Select the table, including its headings but excluding any total row, and choose Insert > Chart.
- In the chart editor’s Setup tab, choose Pie chart. Check that the first column supplies the labels and the second supplies the values.
- Under Customize, edit the title in Chart & axis titles. For the outside labels shown here, set Legend > Position to Labeled. Alternatively, choose Pie chart > Slice label > Percentage to place percentages on the slices, and keep a separate legend for the category names.
Keep the source table beside the chart while editing. If you change a request count in that table, the chart recalculates the shares. Changing a category name in the table also keeps the label tied to the data.
How to make a pie chart in Python
With Matplotlib installed, these six lines create the same chart. You can run them in a Deepnote Python block, keeping the values and the resulting visualization together.
import matplotlib.pyplot as plt
labels = ["Account access", "Billing", "Product questions", "Other"]
counts = [80, 60, 40, 20]
plt.pie(counts, labels=labels, autopct="%1.0f%%", startangle=90)
plt.title("Customer support requests by topic, April")
plt.show()
The labels list names the slices, and counts controls their size. autopct displays percentages rounded to whole numbers; startangle=90 starts the first slice at the top. Matplotlib calculates each share from the supplied values.
Keep each label paired with its count when editing the lists. For real customer records, summarize the requests by topic first rather than drawing a slice for every individual request.
In Deepnote, Agent can help write and run that preparation step. For example, ask it to “Count requests by primary topic for April, check the total, and use those counts in the pie chart.” The calculation remains available beside the output for review.
How to make a pie chart in PowerPoint
In PowerPoint for the web, open Insert > Chart and choose a pie chart. Select the chart and choose Edit in Excel to open its data and chart preview in another browser tab.
Replace the sample data with the four request topics and counts. Use the chart preview in Excel to adjust the title, colors, legend, and percentage labels. Return to PowerPoint to check the result on the slide.
The split view shows how the two browser tabs work together: Excel holds the editable data and chart preview, while PowerPoint shows the presentation. This lets you adjust the figures and formatting without rebuilding the chart on the slide.
In desktop PowerPoint, choose Insert > Chart > Pie and replace the sample data in the spreadsheet that opens. To change the figures later, select the chart and choose Chart Design > Edit Data.
Before presenting, check the slide at presentation size. Make sure the title, percentages, and category names are readable from a distance.
Which method is fastest for your task?
The shortest route usually depends on where your data already live and what you need after the first chart.
- Start in Deepnote when the chart is part of a larger analysis. Use the maker for a quick chart from prepared data, or use Agent when the records need to be grouped, filtered, checked, or revisited later. The analysis and visualization can stay together as the question develops.
- Stay in Excel or Google Sheets when your data are already in a spreadsheet. Create the chart beside the source table so you can update the values and check the result together. Use PowerPoint when you need the chart in a presentation.
- Use Python when the calculation needs to be repeated or customized. In Deepnote, the code, chart, and explanation can remain in one notebook instead of being passed between separate files.
When a pie chart is the wrong call
A pie chart needs categories that form one meaningful whole. Skip it for overlapping survey responses, negative amounts, or values that describe unrelated totals. When categories have similar sizes, compare them with a bar graph or a Cleveland dot plot, which places one marker at each category’s value. Those formats make small differences easier to judge. Use our guide to pie charts for more on choosing a suitable part-to-whole comparison.
Create, refine, and share your pie chart in Deepnote
In Deepnote, the same workflow can continue beyond the first chart. If your CSV or Excel file contains one row per customer request, Agent can group the records by primary topic, surface missing categories, check the total, and prepare the counts used by the chart. The source data, preparation steps, and result stay together, so you can inspect how each percentage was produced and ask follow-up questions without rebuilding the analysis.
To work with an existing table in your notebook, click Visualize on its output to create a Chart block. Select Pie, use Request topic for the categories, and choose Sum of Requests for slice size. Chart AI can configure the chart from a description, and you can adjust its settings afterward.
Use Sum for this example because each row already contains a topic’s request count. Count would count the four rows, giving each topic an equal slice. Under Value labels, choose percentages of the total and use whole numbers.
To share the finished chart, open Share block, enable sharing, and select the output. Anyone with the link can view that shared output. Copy the link or embed the chart in a page; the embed updates when the notebook output changes. Your team can return to the source data and calculations when the next question comes up.
If this becomes a recurring report, keep the month or team fields in the analysis instead of creating separate copies of the chart. Update the source, rerun the notebook, or schedule it to run on a regular cadence. A published data app can expose the latest result, while the notebook retains the calculations behind it.