Bar graphs Comparison
Bar graphs plot numeric values for categorical data as bars on an x/y axis. The numeric amounts are shown on a common baseline in order to easily compare values across categories. If you need to plot and/or compare data across a time range, use a Line graph instead.
Components
For components, go to:
- Web component: ChartGraph
- Figma component: Column bar graph
- Figma component: Horizontal bar graph
Usage guidelines
- When comparing categorical data sets based on total amount and size
- When a side-by-side visual representation is needed to compare sizes
- When viewing or comparing data and/or trends that change in value over time. Use a line graph instead.
- When it’s more important to see how different amounts of categorical data form a whole. Use a pie or donut chart instead.
Best practices
These are best practices specific to bar graphs. For global best practices, see ChartGraph best practices and General guidelines.
Use to compare total values between categories.
Use to show the progress of values over time. Use a line graph instead.
Use to compare sizes from a common zero baseline to maintain accuracy.
Start from a non-zero baseline, as that can distort the ratios between the amounts being compared.
Use additional colors only when comparing categorical data that sits inside of a larger dimension also being compared. Keep it to 4 colors maximum, otherwise charts will get hard to read.
Use multiple colors to compare only one large category with no sub-categories. Use a single color in those cases as the main focus is comparing amounts.
Variants
See the ChartGraph component for use cases and examples of all the different types of bar graphs available:
Writing
For general writing and content formatting guidelines, refer to our Content standards.
- Keep labels short so that they don’t wrap and make it hard to read data
- Use abbreviations that are commonly understood and can be translated to all supported languages
- Create extra-long labels that have to wrap or truncate
- Use abbreviations that are only understood internally or that don't translate well.