Create an effective bar chart
Create an effective bar chart
1. Confirm that a bar chart is the most appropriate chart type for the data.
Bar charts are best for comparing a metric across categories: If the chart is showing a metric trended over time, consider a line chart instead. Data that can be represented in a pie chart is generally more readable if presented as a horizontal bar chart.
2. Determine whether the chart should be a horizontal bar chart or a vertical bar chart.
A horizontal bar chart is generally more readable than a vertical bar (column) chart, as vertical bar charts can force axis labels to wrap to multiple lines, which can look crowded and be more difficult to scan and interpret. Reading categories from top to bottom, rather than left to right, is more natural and easier on the brain. For example: A vertical bar chart with some category text wrapping to two lines The same data as a horizontal bar chart is easier to read and does not require wrapping However, a vertical bar chart can be appropriate if: There are fewer than four categories included on the chart. The names of the categories are all short (so they do not have to wrap across multiple lines). The space available for the chart in the dashboard, report, or presentation works better as a vertical bar rather than a horizontal one.
3. Use a data visualization platform like Power BI or Tableau and create the basic chart using the platform's default settings.
4. Use neutral colors like gray or light tones as base colors for the bars to improve readability.
Use base colors that align with the palette in your corporate style guide if you have one. Work with one of the default palettes available in your data visualization platform if you don’t.
5. Use vivid colors to highlight any specific bars you want to call out to help them stand out.
For example, highlighting the Non-Branded Paid Search bar in the chart below using a vivid color helps it stand out and draws attention to it.
6. Remove unnecessary visual elements and use lighter colors and tones for non-data elements to improve readability.
Remove the following unless they are absolutely necessary to interpret the chart: Chart borders. Vertical and horizontal grid lines. Axis tick marks. Axis lines. Axis titles (if the nature of the data or the chart title makes them redundant). For the remaining non-data elements: Change axis labels and titles to lighter colors, such as medium gray rather than black. Use light colors, like a very light gray, for any remaining grid lines. Before removing unnecessary elements After removing unnecessary elements
7. Create multiple charts with categories that share the same starting point instead of using stacked bar charts.
Stacked bar charts are difficult to perform comparisons on for all categories beyond the first category because they do not have an aligned starting or ending point. Comparing Desktop and Tablet across categories is difficult The following example shows the same chart separated into three separate charts: Three separate charts instead of a stacked bar chart