How to Create a Donut Chart in Tableau

Cody Schneider

Building a donut chart in Tableau feels like you’re using a secret trick, but it's a straightforward skill that can make your dashboards more engaging and informative. This chart type is essentially a pie chart with the center cut out, which you can then use to display key information like a grand total. This tutorial will walk you through creating a donut chart from scratch, step-by-step.

What Is a Donut Chart, and When Should You Use One?

A donut chart displays data in slices, just like a pie chart, with each slice representing a category's proportion of the whole. The primary difference is the hole in the center. This design choice isn't just for looks, it addresses some of the critiques of a standard pie chart. By removing the center, you encourage the viewer's eye to compare the length of the arcs on the circumference rather than the angles of the slices, which can be easier to interpret accurately.

The empty space in the middle is also prime real estate for adding context. You can place a total sum, an average, or another key metric inside the chart, giving your audience a main takeaway at a glance while also showing the proportional breakdown.

Use a donut chart when:

  • You need to show how individual parts make up a whole.

  • You have a small number of categories (ideally fewer than six) to avoid a cluttered and unreadable chart.

  • You want to display a key total or KPI prominently in the center of the chart.

Examples include showing sales distribution by product category, website traffic sources as a percentage of the total, or budget allocation across departments.

Avoid using a donut chart when:

  • You have too many categories, which makes the arcs too thin to compare.

  • The values of your slices are very close to each other, making visual comparison difficult.

  • You need to compare data across multiple groups. A bar chart is almost always better for comparing values against each other.

Preparing Your Data for a Donut Chart

The good news is that donut charts don't require a complex data structure. To build one, you typically only need two fields from your data source:

  • A categorical dimension: This field separates your data into slices. Examples include Product Category, Region, Marketing Channel, or Lead Status.

  • A numerical measure: This field determines the size of each slice. Examples include Sales, Profit, User Count, or Revenue.

For this tutorial, let’s imagine we're using the Sample - Superstore dataset that comes with Tableau. We'll use the 'Category' dimension and the 'Sales' measure.

Here’s what our data conceptually looks like:

Category SalesFurniture $741,999.86Office Supplies $719,099.64Technology $836,155.15

Step-by-Step Guide to Creating a Donut Chart in Tableau

Tableau doesn't have a one-click "Donut Chart" option in its Show Me panel. Instead, we create it by building a pie chart, layering a blank circle on top of it, and then styling the final product. Let's get started.

Step 1: Create a Basic Pie Chart

First, we need to create a simple pie chart. This will form the outer ring of our eventual donut.

  1. Connect to your data source (e.g., Sample - Superstore).

  2. In the Marks card, click the dropdown menu and change the mark type from Automatic to Pie.

  3. Drag your dimension (e.g., Category) onto the Color mark.

  4. Drag your measure (e.g., Sales) onto the Angle mark.

You should now see a pie chart where each colored slice represents a product category, and the size of the slice is determined by its total sales. You can resize the chart by holding Ctrl + Shift + B (to make it bigger) or Ctrl + B (to make it smaller).

Step 2: Create the "Hole" for the Donut

This is where the magic happens. To create the hole, we’ll use a clever trick involving a dual-axis chart. We will create two identical pie charts in the same view and then lay a smaller, plain white circle over the second one.

  1. In the Rows shelf, double-click an empty area. This will open a small text box.

  2. Type in AVG(0) and press Enter. You'll see a pill appear on the Rows shelf. This creates a quantitative axis, and your pie chart might turn into a circle at the zero mark.

  3. Drag this new AGG(AVG(0)) pill while holding the Ctrl key (or Command on Mac) to duplicate it right next to the original on the Rows shelf.

You will now see two identical pie charts in your view, one on top of the other.

Step 3: Combine the Charts Using a Dual Axis

Now, we need to merge these two pie charts into a single space so we can layer one on top of the other.

  1. Right-click the second AGG(AVG(0)) pill on the Rows shelf.

  2. From the context menu, select Dual Axis.

Your two pie charts will now be overlaid, appearing as one. Notice that your Marks card has changed. You now have three sections you can edit: All, AGG(AVG(0)), and AGG(AVG(0)) (2). These correspond to the overall view, the first pie chart, and the second pie chart, respectively.

Step 4: Format the Hole

Our next goal is to turn the top pie chart into a solid-colored circle that will become our "hole."

  1. In the Marks card, select the second chart editor, labeled AGG(AVG(0)) (2).

  2. Remove Category from Color and Sales from Angle by dragging them off the card. The chart will likely turn into a single gray circle.

  3. Click on the Color mark and choose white (or whatever background color your dashboard has).

  4. Click on the Size mark and drag the slider to the left to make this white circle slightly smaller than the pie chart behind it.

Voila! You now have a donut chart. The colored pie chart from the first Marks card is visible around the edges of the smaller, solid white circle you just created.

Step 5: Add a Total Value to the Center

An empty hole isn't very helpful, so let's use that space to display the total sales for all categories.

  1. Make sure you still have the AGG(AVG(0)) (2) Marks card selected (this is the card for our white circle).

  2. Drag your measure (Sales) onto the Label mark.

You'll now see the sum of sales displayed right in the middle of your donut chart. You can format this label further by clicking on the Label mark, which allows you to adjust the font, size, and color to make it stand out.

Step 6: Final Cleanup and Formatting

Your chart looks great, but there are a few lingering elements from the dual-axis setup that we should clean up.

  1. Remove Headers: Right-click on the "AVG(0)" axis on the left side of the chart and uncheck Show Header.

  2. Remove Zero Lines: Navigate to the top menu and click Format > Lines. In the Format pane on the left, go to the Sheet tab, find Zero Lines, and select None. This removes the faint gray line running through the middle of the chart.

  3. Add Slice Labels: Select the first Marks card (AGG(AVG(0))) and drag Category onto the Label mark. Then drag Sales onto the Label mark as well.

  4. Format Slice Labels: Right-click the SUM(Sales) pill within the Marks card and select Quick Table Calculation > Percent of Total. Now your slices will show the category name and its percentage of total sales. Click the Label mark to further customize the text alignment and format.

After these steps, you have a polished, presentation-ready donut chart that effectively communicates proportions while highlighting the most important overall metric.

Final Thoughts

Creating a donut chart in Tableau is an excellent example of using the tool's flexibility to build visualizations that go beyond the default options. By combining two charts on a dual axis, you can turn a standard pie chart into a more modern and information-dense donut chart that clearly shows both parts of a whole and the total value.

While mastering tools like Tableau is incredibly rewarding, there are times you need answers fast without the manual clicks. We created Graphed for exactly that purpose. You can simply connect your data sources - like Google Analytics, Shopify, or Salesforce - and then ask for the chart you need in plain English. For example, you could ask, "Show me sales by product category as a donut chart for last quarter," and get an interactive visualization in seconds, freeing you up to focus on the insights, not just the setup.