The Doughnut Chart is a built-in chart type in Excel. Doughnut charts are meant to express a “part-to-whole” relationship, where all pieces together represent 100%. Doughnut charts work best to display data with a small number of categories (2-5). For example, you could use a doughnut chart to plot survey questions with a small number of answers, data split by gender, Windows vs. Mac users, or other data where categories are limited.

Doughnut charts should be avoided when there are many categories, or when categories do not sum to 100%.

Pros

  • Simple, compact presentation
  • Can be read “at a glance”
  • Excel will calculate percentages automatically

Cons

  • Difficult to compare relative size of slices
  • Become cluttered and dense as categories are added
  • Limited to part-to-whole data
  • Poor at showing change over time

Tips

  • Limit categories
  • Consider other charts to show change over time

The Pie of Pie Chart is a built-in chart type in Excel. Pie charts are meant to express a “part to whole” relationship, where all pieces together represent 100%.

Pie charts work best to display data with a small number of categories (2-5). The Pie of Pie Chart provides a way to add additional categories to a pie chart without generating a pie chart too complex to read. When configuring a Pie of Pie chart, Excel provides a setting that moves the smallest n slices of the pie to another smaller pie, where n can be adjusted to suit the data.

Pie of pie charts should be avoided when there are many categories, or when categories do not total 100%. The human eye has trouble comparing the relative size of slices in a pie chart, and this problem is compounded with the pie of pie variety.

Pros

  • A simple way to handle more categories in a pie chart
  • Can be read “at a glance” with limited categories
  • Excel can calculate and display percentages automatically as data labels

Cons

  • Difficult to compare relative size of slices
  • Become cluttered and dense as categories are added
  • Limited to part-to-whole data
  • Poor at showing change over time

Tips

  • Limit categories
  • Avoid all 3d variants