What is a Pie Chart and How is it Used in Astrato?

Pie charts might be the most misused data visualization of all time. Percentages adding up to 137%, a thousand different categories, disturbing color palettes. But these are all user errors. The pie chart itself is an incredibly powerful visualization tool that can deliver invaluable Business Intelligence for your team and organization. In this article, we explore how to (correctly, and effectively!) use pie charts in Astrato.

Pie charts are used to show the proportion of each category in a data set. For example: To understand the relative size of departments in a company, individual segments of a pie chart can be used to visualize the size of the sales team or product department, etc., as a percentage of the size of the company.  

And if you wanted to know what brands of phone the employees are using, you could have a slice for iPhone, Samsung, Nokia, etc. 

The key feature in these examples is that we are looking at members of a specific population (i.e. the company)  and then segmenting them by different categories (i.e. department, brand of phone). 

Any time you want to express a part-to-whole relationship in your data, you should consider a pie chart.  

Best Practice for Pie Charts

As with any chart, there is no one-size-fits-all set of rules that governs every set of data. But there are some principles that you can follow for making an effective pie chart. They are:

1. Limit the Number of Segments:

Too many slices can crowd the chart and make it difficult to read. We recommend between two and five for clarity. A donut chart is an alternative when comparing fewer variables (only two or three). The benefit of a donut chart is that the white space in the middle can be used to communicate additional information  (e.g. the sum of the categories equals an increase in revenue). 

Good (Few segments):

Bad (Too many segments):

2. Order the Slices From Largest to Smallest:

People are generally interested in which category is the biggest and which is the smallest. Ordering by size communicates this best. If you imagine the pie chart as a clock, starting with the biggest slice at 12 o’clock and moving clockwise – from largest to smallest – you’ll be able to communicate relative size most effectively. 

Good (Logical flow from 12 o’clock):

Bad (Segments appear randomly ordered):

3. Use Data Labels and Annotate:

Knowing which segment is which is essential, whether it’s with data labels or a legend. Further annotation can provide useful context or additional information, how proportions have changed over time for example, and help communicate the broader message of the chart.

Good (Clear, concise labels):

Bad (No information about what the segments represent):

4. Are you sure you want a pie chart? 

There’s rarely one way to visualize a dataset. It can be worth exploring other options to make sure you have found the one best suited to your data and audience. A couple of the most common pie chart alternatives are stacked bar charts and treemaps. 

How to Build a Pie Chart in Astrato

Making pie charts in Astrato is quick and intuitive, just follow these steps:

1. Within the Visualization section, click the Pie object icon, and then drag and drop the chart icon to your worksheet.
2. Use the data tab in the object panel to select a dimension and measure you would like to visualize with the pie chart.

3. Customize the properties of the pie chart with the Style and Layout options. Style your pie chart by customizing your color palette, adding borders, and showing or hiding the legend.

How Different Industries Use Pie Charts

Understanding the size of different categories relative to the larger group has a range of use cases, so pie charts are deployed in a range of industries. Here are five examples:

Healthcare and Medicine

In healthcare, pie charts can be used to deliver insights about a hospital or the broader population. Some common use cases are the prevalence of different diseases, patient demographics, and the allocation of healthcare resources. The European Centre for Disease Prevention and Control uses this pie chart to show healthcare-associated infections.

Google Analytics

As in this demo Astrato workbook, pie charts can be used to present what is driving people to a website. The chart updates as filters are added and removed, allowing the user to drill down into the data. In this example, we learn that Google Chrome users are more likely to arrive via referral than the average of other browsers. A powerful insight, delivered by an Astrato pie chart.

Census Data

Possibly the most common use of pie charts is for demographic and census data. This might be vote share following an election or more general population information like age group and level of education. Pew Research Center used a pie chart to visualize the relative size of major religions around the world. 

Inventory or budget tracker

Pie charts are a useful way of tracking inventory status and budget allocation. Here you can see remaining stock, visualized in a pie chart, alongside how much of each product has been shipped. 

Market research

Market share might be the most common way pie charts are utilized in marketing, but they are also useful for understanding target audience distribution or broader customer demographics. Pie charts can represent data related to customer preferences, buying patterns, or advertising budget allocation across different channels. 

Summary

Use pie charts to summarize part-to-whole relationships in your data, don’t use too many segments, order them by size, and label them! Ready to use pie charts in your BI dashboard? You can find out more about pie charts and Astrato’s other visualizations here.