Colour is one of the most powerful tools in data visualization — and one of the most frequently misused. Used purposefully, colour guides attention, groups related elements, and encodes meaning. Used carelessly, it creates confusion, excludes readers, and obscures the data.
Use Colour with Purpose
Every colour in a chart should serve a reason. Common purposes include:
- Emphasis — highlighting one bar or line that is the focus of the finding, while keeping others neutral.
- Grouping — using the same colour for related categories.
- Encoding a value — using colour intensity to represent a quantity (sequential or diverging scales).
Decorative colour — using many colours simply to make a chart look varied — adds visual noise without adding information.
Colour Schemes
Categorical (qualitative) schemes use distinct colours for unordered categories. Each colour should be clearly different from the others. Limit the number of categories to what can be distinguished reliably — typically six or fewer distinct colours.
Sequential schemes use a gradient from light to dark (or low to high saturation) to represent ordered values. Useful for maps and heatmaps where higher values should appear more intense.
Diverging schemes use two contrasting colours with a neutral midpoint. Useful when values range from negative to positive, or when a meaningful midpoint exists (such as a target or average).
Colour Vision Differences
Approximately 8% of men and 0.5% of women have some form of colour-vision difference. The most common form affects the ability to distinguish red from green. A chart that relies solely on red and green to convey meaning will be unreadable for a significant portion of your audience.
Design for colour-vision accessibility by:
- Using redundant cues — combine colour with shape, pattern, or direct labels so that colour is not the only way to distinguish categories.
- Choosing colour-vision-friendly palettes — blue and orange, or blue and red, are more distinguishable than red and green for most colour-vision differences.
- Testing your chart in greyscale — if the chart is still readable in greyscale, it will likely work for most colour-vision differences.
Designing for colour-vision accessibility also improves readability for people viewing charts on low-quality screens, in bright sunlight, or in printed black-and-white. Redundant cues and sufficient contrast are good practice for all audiences.
Contrast
Text and labels must have sufficient contrast against their background to be readable. This applies to axis labels, data labels, legend text, and annotations. Low-contrast grey text on a white background may look elegant but fails accessibility standards and becomes unreadable when printed or projected.
The same contrast requirements apply to interactive dashboards and web-based charts, not only to static images.
Colour and Cultural Context
Colour carries cultural associations that vary by context and audience. Red often implies danger or loss in financial contexts; green often implies positive or safe. These associations are not universal. When colour carries implied meaning, label the chart clearly so that the meaning does not depend on cultural assumptions alone.
Limiting Colour Count
More colours do not create more clarity. A chart with ten different colours is harder to read than one with three. When a chart requires many categories, consider grouping minor categories, using a table instead, or splitting the chart into smaller focused views.
Common Mistakes
- Using red and green as the only distinguishing factor between two categories.
- Using many colours decoratively without encoding meaning.
- Low-contrast labels that are difficult to read.
- Assuming a colour palette is universally accessible without testing.
- Relying on colour alone to convey information that screen readers or greyscale printing cannot reproduce.
Key Takeaways
- Every colour should serve a purpose: emphasis, grouping, or value encoding.
- Use redundant cues — colour plus shape, pattern, or label — so colour is not the only distinguishing factor.
- Test charts in greyscale to check colour-vision accessibility.
- Ensure sufficient contrast for all text and labels.
- Limit the number of distinct colours to what can be reliably distinguished.
- colour-blind-safe data visualization
- accessible chart colour design
- sequential diverging categorical colour schemes
- chart contrast accessibility standards