Choosing Colors for Charts and Data Visualizations

Chart color is a specialized domain with rules that conflict with standard brand color thinking. A vivid brand palette built for a marketing site will likely fail in a dashboard context. **The Three Palette Types** Data visualization uses three distinct color structures, each for different data types: *Categorical palettes* encode unordered groups (regions, product lines, categories). Colors should be maximally distinct — different hue, similar lightness. If one category color is much darker or lighter than others, it will appear more important, distorting the data's message. Target 6-8 maximally distinct colors; beyond that, consider labels or patterns. *Sequential palettes* encode ordered, continuous data (temperature, revenue, age). Use a single hue progressing from light (low value) to dark (high value), or a perceptually uniform gradient. Avoid rainbow sequences (jet, spectrum) — they introduce false boundaries and are systematically inaccessible to red-green colorblind users. *Diverging palettes* encode data with a meaningful midpoint (profit/loss, positive/negative, above/below average). Use two hues diverging from a neutral center — typically a pale gray or white. Saturation should increase as values move away from center in either direction. **Why Brand Colors Fail in Charts** Brand palettes are typically constructed for hierarchy (primary, secondary, accent) not distinction. Using your brand blue, brand orange, and brand teal in a categorical chart creates instant confusion: which is 'more important'? Chart colors need equal perceptual weight, not hierarchy. **Colorblind Testing** Approximately 8% of males and 0.5% of females have some form of color vision deficiency. Deuteranopia (red-green) is most common. Before finalizing any categorical palette, simulate it in deuteranopia mode (our colorblind tool works for this). If categories become indistinguishable, add pattern, shape, or label redundancy. Never rely on color alone to encode categorical data.