Pie Chart

Pie chart divides a circle into labeled slices proportional to each category's share of the total. Use it only when composition, not comparison, is the primary message, and when you have five or fewer categories.

Figure purpose

Pie chart encodes part-to-whole relationships using angle and arc length. Its strength is communicating rough shares of a total at a glance, for example, that one category accounts for roughly half, and two others account for roughly a quarter each. It is not suited for tasks that require readers to compare nearby values accurately.

When to use

Use pie chart when all of the following are true:

  • You have five or fewer categories. With more, the slices become difficult to distinguish and the legend grows harder to scan.
  • The differences between your largest and smallest slices are meaningful; a rough guide is that adjacent slices should differ by at least 10 percentage points to be visually distinguishable.
  • The question is "what share of the total?" not "how do these values compare?"
  • You do not need readers to identify exact percentages; use a table for that.

Licklider cannot determine automatically whether your values are semantically a part-to-whole composition or whether small slice differences will still be interpretable for your audience. It can count categories and render percentages, but it cannot fully judge whether the real scientific question is composition rather than comparison.

Do not use pie chart when:

  • Slices are similar in size (the chart will not reveal the difference reliably)
  • You have more than five categories
  • The real question is a group comparison; use a strip plot or box plot instead
  • You need to show change over time; use a line chart

Required columns

Pie chart requires exactly two columns:

  • A category label column (categorical): one row per category, or multiple rows that Licklider will aggregate by summing values within each label.
  • A numeric value column: the size of each slice. Values do not need to sum to 100; Licklider calculates percentages automatically.

Rows with missing labels or non-numeric values are silently excluded. Zero-value categories are included as zero-width slices.

What Licklider displays

Slice labels are configurable. Licklider can display Name + %, Name + value, Name + value + %, % only, or Value only directly on the slices. Hovering over a slice also reveals the exact value alongside the label and percentage.

Licklider's figsafe system will warn or block rendering when the design has 12 or more categories. For best results, keep category count to 5 or fewer.

This guard mainly addresses obvious structural overload such as too many slices. It does not guarantee that the chart is the best semantic choice for the message. A pie chart can still be a poor fit when the real task is precise comparison rather than rough composition reading.

Example

Appropriate use

A clinical trial reports the distribution of primary outcome categories: Complete response (42%), Partial response (31%), Stable disease (18%), Progressive disease (9%). Four categories, clear differences, composition is the message.

Inappropriate use

Five treatment groups with mean response values of 12.1, 13.4, 11.8, 14.2, and 12.9. These values are close, they represent means not shares, and the question is comparison not composition. Use a strip plot or box plot.

Alternative figures

  • Use Donut Chart when the same message would benefit from a center label showing total n, total value, the largest category, or custom text.
  • Use Stacked Bar Chart when composition must be compared across several groups instead of summarized for only one whole.
  • Use a bar chart or Strip Plot when readers need to compare values accurately across categories.
  • Use Group Comparison figures when the underlying question is whether groups differ, not how a total is composed.

Design Rationale & References

Pie charts are treated as use-with-caution figures because angle and arc length support rough part-to-whole reading but are weaker than position or length for precise comparisons. That is why Licklider recommends them only for small category counts, discourages use when slices are similar, and routes comparison-heavy questions toward bar-like or point-based figures instead [1, 2].

The guidance to keep pie charts to five or fewer categories and to prefer clearly separated slices is also intentional. As category count rises, labels and legend scanning become harder, and perceptual comparison between slices degrades. The current figsafe warning for very high category count is a guardrail against obvious overload, not a claim that all smaller pies are automatically effective.

For the broader perceptual basis and neighboring part-to-whole figure choices, see Part to Whole - Design Rationale & References.

  1. Cleveland, W. S., & McGill, R. (1984). Graphical perception: Theory, experimentation, and application to the development of graphical methods. Journal of the American Statistical Association, 79(387), 531-554. https://doi.org/10.1080/01621459.1984.10478080
  2. Heer, J., & Bostock, M. (2010). Crowdsourcing graphical perception: Using mechanical turk to assess visualization design. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 203-212. https://doi.org/10.1145/1753326.1753357

See also