Data Visualization Color Guide: Designing for Accessibility (Stop Using Red & Green)

Data Visualization Color Guide: Designing for Accessibility

Color is not just decoration; it is data. If you choose the wrong colors, you aren't just making an ugly chart—you are actively hiding insights from your audience.

Most beginners design dashboards based on "what looks pretty." They pick the corporate brand colors, or they default to the classic "Traffic Light" (Red, Yellow, Green) palette.

This is a mistake. Approximately 1 in 12 men (8%) and 1 in 200 women suffer from some form of Color Blindness (Color Vision Deficiency). The most common form is Deuteranopia, which makes Red and Green look exactly the same (a muddy brown-yellow).

If your KPI dashboard uses Red for "Bad" and Green for "Good," 8% of your audience literally cannot tell if the company is making money or losing it.

The Golden Rule: Never rely on color alone to convey meaning. Use icons, labels, or position combined with color. And stop using Red/Green for comparisons. Use Blue/Orange instead.

The 3 Types of Data Palettes

To design professionally, you must understand that different data requires different colors. There are three main categories:

1. Categorical Palettes (Qualitative)

Use for: Distinct items that have no inherent order (e.g., Departments, Product Categories, Countries).

Strategy: Colors should be distinct and have the same "visual weight." Don't make "Sales" a bright neon green and "Marketing" a dull grey, or people will think Sales is more important.

Good Example: Blue, Orange, Purple, Teal. Bad Example: Light Blue, Medium Blue, Dark Blue (Implies order).

2. Sequential Palettes

Use for: Data that goes from Low to High (e.g., Revenue, Temperature, Age).

Strategy: Use a single hue that varies in lightness/saturation. Light usually means "Low" and Dark means "High."

Example: White -> Light Blue -> Medium Blue -> Dark Navy.

3. Diverging Palettes

Use for: Data that has a meaningful midpoint (e.g., Profit vs Loss, Likes vs Dislikes, Vote Share).

Strategy: Two contrasting hues meet at a neutral middle (usually white or grey). This allows the user to instantly see "Positive" vs "Negative."

Safe Example: Dark Orange -> White -> Dark Blue.

Tools for Testing Accessibility

You don't have to guess. There are free tools that simulate how your dashboard looks to colorblind users.

• Adobe Color (Accessibility Tools): Lets you create palettes and flags conflicts instantly.
• Coblis (Color Blindness Simulator): Upload a screenshot of your dashboard to see it through different eyes.
• Power BI / Tableau: Both have built-in "High Contrast" or "Color Blind Safe" themes. Use them.

The Psychology of Color (Cultural Context)

Beyond biology, there is psychology. Colors have meaning, but that meaning changes based on context.

Color	Western Business Context	Potential Pitfall
Red	Danger, Stop, Loss, Hot	In China, Red means Good/Luck/Up.
Green	Good, Go, Profit, Eco	Indistinguishable from Red for many.
Blue	Trust, Stability, Cold	Can feel unemotional or corporate.
Grey	Neutral, Past Data, Context	Too much grey looks "disabled."

Pro Tip: Use Grey for context. If you want to highlight "This Year's Sales," make this year Blue and make "Last Year's Sales" Grey. This is called "Preattentive Processing"—it forces the eye to look where you want.

Conclusion

Data visualization is about communication. If your color choices create a barrier to understanding, you have failed as an analyst. Designing for accessibility doesn't make your work "boring"; it makes it universal.

Stop painting with the whole rainbow. Pick a safe palette and stick to it.

Download January Skills: Color Blind Safe Palettes (HEX Codes)