Pie Charts and Donut Charts are familiar, yet powerful data visualization tools that play a crucial role in Sigma Computing, a platform renowned for its robust data analytics capabilities. These charts are instrumental in transforming complex datasets into easily digestible and visually appealing representations for most part-of-whole analytical use cases
What are Pie and Donut Charts?
A Pie Chart is a circular graphical representation of data that is divided into segments or “slices” where the total area of each slice and the center angles illustrate proportionality.
A Donut Chart is a type of data visualization that is similar to a Pie Chart, but with a hole or “doughnut” in the center, and it displays categories as arcs rather than slices. These are most commonly used to display % to total for a single metric, such as the goals on your wearable fitness tracker.
What are the Benefits of Pie and Donut Charts?
As previously stated, these are familiar visuals for users and can be powerful aspects of your data narrative when used for a part-to-whole analysis. While there are exceptions to almost every rule in design, Pie and Donut charts are decidedly ineffective for most other analytical concepts such as distribution, ranking, deviation, flow, correlation, and change over time.
We will cover some important configuration and design best practices later, but limiting their application to just part-of-whole use cases will help you avoid creating a confusing chart.
How to Create a Pie Chart in Sigma
We are going to use the D_CUSTOMER table from PLUGS_ELECTRONICS in the Sigma Sample Database if you’d like to follow along. These steps begin after a data element exists in your workbook.
To learn how to add data to a workbook, check out How to Create a Data Element in Sigma.
Step 1: Add a new Viz Element
Once in Edit mode in your Workbook, click the plus sign In the top left corner of the Sigma Workbook interface.
The below interface appears in which you need to select the Viz under the DATA ELEMENTS pane.
Next, you will be prompted to select a source for the Viz element. Click on the Page Elements tab and select the
D_CUSTOMERtable element. (If you don’t see the data you want to use as a source, make sure it has been added to this workbook:
How to Create a Data Element in Sigma)
Now you can see the pane where you need to select the viz type. By default, the bar chart is selected, so in the visualization pane, click the drop-down menu and select Pie.
Step 2: Build the Pie Chart Visualization
Here we are going to create a Pie Chart for Total Customers By Customer Loyalty Program Membership. Loyalty Program membership is recorded as 1 for a member and 0 for a non-member.
To make this easier to understand, you can add this calculation to your table:
If([Loyalty Program] = "1", "Member", "Non-Member")
Click the plus sign (“+”) on the color selection, type in “Loyalty Program Membership,” and click on the column name. You can also drag and drop the column name from the column list onto the Color section.
Next, we are going to add the count of Customer Names.
In the Value Section, click the plus sign, and type in “Cust Name,” and click on the column name. You can also drag and drop the column name from the column list onto the Value section.
By default, Sigma will take the count since this is a text field. If the measure you’d like to display is already aggregated to the level on your color field, you can uncheck the Aggregate Values.
Your Pie Chart should look like this.
-
(Optional) Change slice colors
-
If you’d like to change the colors of each slice, click the downward-facing arrow in the Color section to either select a different palette or assign custom colors.
-
Step 3: Add Labels and Adjust Other Formatting Options
We always recommend using three or fewer categories on Pie and Donut Charts because it is difficult for the brain to correctly compare curved slice sizes. No matter how many slices are in your Pie or Donut Chart, it is always essential to include data labels so the end user can correctly interpret the data.
To add labels and adjust other formatting options, click on the paintbrush icon to go to the element format menu. Here are the current formatting options available:
BACKGROUND – Click on the BACKGROUND section to show its options. Here in Show background tick the checkbox to show the Background color option to change the background of your visualization canvas.
-
TITLE – Click on the TITLE section and by default, it is tick-marked to enable you to change the title of the visualization. You can also directly click on the title above the visualization to rename it. Lastly, you can change the font boldness, color, size, and alignment. We renamed our title to “Customer Loyalty Membership”.
-
Note: Once you change the title of an element, Sigma no longer automatically updates the title. For example, if we changed the measure of this Pie Chart to be a count of orders by membership type, the title would not automatically update to reflect this change. If we had not changed the title or “Revert to default” on the title, Sigma would automatically change the title from “Count of Cust name by Membership” to “Count of Order ID by Membership”.
-
LEGEND – Click on the LEGEND section to change the Position, Label font size, Label text color, and Show legend header. In our case, we set values in the legend section as shown below.
-
DATA LABELS – Click on the DATA LABELS to change what is displayed, Label style, Percent decimal precision, Color, and size of Font. In our case, we set values in the Data Label section as shown below.
How to Create a Donut Chart in Sigma
To create a Donut Chart in Sigma, follow the same steps as listed above, except for when you add a new Viz element, select Donut instead of Pie (Step 1.4).
If you already have a visual built that you’d like to convert into a Donut Chart, you can change it to a Donut Chart by clicking on the Visualization Type drop-down menu and choosing Donut.
Best Practices for Designing Pie and Donut Charts
Limit the Number of Categories
Avoid creating charts with too many categories. Like trying to find the biggest and smallest slice of a party-sized pizza, judging the slight differences in area and the center angles of a dozen or more segments is extremely difficult for the human eye. Pie and Donut Charts are most effective when displaying no more than three segments. Too many segments will also impede your ability to fit labels.Always Use Labels
Add labels to each segment, displaying at minimum the number or percentage and typically the category name. Ensure that labels are legible and don’t overlap. There are additional format options for the placement, order, and size of color legends. If space is very limited, consider using a text box to replace the legend and use a “legend-in-text” title.
Sort Categories
Arrange the categories in a logical order, such as from largest to smallest or in alphabetical order, to help viewers quickly grasp the distribution.
Interested in best practices for other visuals?
Check out our Best Practices for Sigma Scatter Plots and Bar Charts Blogs!
Closing
As you dive deeper into Sigma and explore the versatility of pie and Donut Charts, remember that the key to effective data visualization lies in both the choice of chart and the way you present it. Be mindful of your data, your audience, and the story you want to tell. These visualizations are not just colorful circles; they are powerful tools that can distill complex data into easily digestible insights, helping you make informed decisions and tell compelling data stories.
Interested in learning more about
Sigma Computing?
Explore our blog for more insightful articles, expert tips, and actionable advice!
FAQs
Pie and Donut charts are divided into segments or "slices", each representing a specific category or data point within the dataset. These segments are arranged radially around the center of the chart and are proportionally sized based on the values or percentages they represent relative to the total.
Pie and Donut Charts are best kept to three or fewer categories. Anything more makes it difficult to quickly and accurately differentiate segment sizes. If your analysis requires more than three categories consider choosing a different visualization such as a bar chart. If a different visualization type is not an option, be sure to have appropriate labels and consider pairing the visual with either another visual type and/or a table to more easily identify the correct numbers.
