Data Visualization

Search intent: data visualization color palette chart color scheme categorical sequential diverging dashboard

Color for Data Visualization: Building Perceptually Correct Chart Palettes

Data visualization color is not brand color. The goal is accurate encoding, not aesthetic harmony. This guide covers the four semantic color roles in dataviz, perceptual uniformity requirements, color blindness constraints, and how to build categorical, sequential, and diverging palettes that communicate data without misleading it.

Data VisualizationChartsColor Theory

Key points

The four semantic color roles in dataviz: categorical (group membership — must be maximally distinguishable), sequential (ordered quantity — must increase uniformly in lightness), diverging (deviation from a midpoint — must be balanced), and highlight (selected/anomalous values). Confusing these roles is the most common chart color failure.

Sequential and diverging scales must use perceptually uniform interpolation (OKLCH or CIELAB). HSL interpolation for gradients produces visible brightness bands and false perceptual midpoints — a well-documented failure mode that makes charts misleading.

Approximately 8% of men have red-green color vision deficiency. Categorical dataviz palettes must maintain distinguishability through lightness variation alone — color cannot be the sole encoding signal.

Building a production-ready categorical palette

A categorical palette for dashboards requires: (1) Minimum 15 OKLCH L units of lightness difference between any two adjacent colors — this ensures grayscale distinguishability. (2) Temperature alternation — alternate warm and cool hues to prevent perceived gradient ordering between unordered categories. (3) Maximum 8 categories per chart. Beyond 8, group small categories as 'Other.' (4) Semantic reservation — designate specific hues for semantic roles: a warm amber is always 'warning,' a red is always 'negative,' a green is always 'positive.' Never assign these hues to neutral data categories. The full ColorArchive provides a useful starting palette: select one medium-lightness, medium-saturation color from each color family, then check that each pair passes the lightness test.

Sequential and diverging scale construction

Sequential scales (for ordered data: quantity, intensity, risk level): (1) Choose a single hue. (2) Create an 7-9 step scale from near-white to a dark, saturated value of that hue. (3) Verify the scale is monotonically increasing in perceived lightness by converting each step to grayscale — if any step reads lighter than the previous, the scale is not perceptually monotonic. Diverging scales (for deviation from a midpoint: positive/negative, above/below average): (1) Choose two complementary hues — one for positive deviation, one for negative. (2) Ensure the midpoint is a neutral light gray. (3) Both sides must have equal maximum saturation and equal minimum lightness at their extremes. A common failure: using green/red as the diverging pair — this combination is impaired for roughly 8% of the male population. Prefer blue/orange for the primary diverging pair.

Practical next step

Move from the guide into a concrete palette lane

Guides explain the use case. Collections prove the taste. Packs handle the export and implementation layer.

Browse packs More guides

Related guides

Data Design

Color in Data Visualization: Sequential, Diverging, and Categorical Palettes

Data visualization is one of the highest-stakes environments for color decision-making. The wrong palette can make a chart misleading, inaccessible to colorblind viewers, or unreadable in grayscale. This guide covers the three palette types, how to validate them, and the most common visualization color mistakes.

Data Visualization

Color in Data Visualization: Sequential, Categorical, and Diverging Palettes Explained

Data visualization color follows a different set of rules from brand or UI color — the palette must encode information accurately, survive colorblindness, and remain legible at small scales and in print. This guide covers the three palette types every data designer needs, how to build each, and the most common mistakes that make charts misleading.

Accessibility

Color Accessibility in 2025: WCAG 2.1, APCA, and the Dual-Standard Audit

WCAG 2.1's contrast ratio formula was not calibrated for font weight or size — it treats a 400-weight 12px label and a 700-weight 48px heading identically. APCA (Advanced Perceptual Contrast Algorithm), the foundation of the forthcoming WCAG 3.0, corrects this with a model that varies required contrast by font size and weight. This guide explains the practical differences, how to run a dual-standard audit, and what to prioritize during the standards transition.