The Importance of Information Visualization
A FuguUX blog post (by Ziyan Tan, May 1 2026) arguing that a well-designed visualization communicates data faster and more memorably than prose or a raw table, because it lets a reader exploit fast, automatic (“pre-attentive”) visual processing rather than parsing text line by line.
License: FuguUX’s own published content — proprietary, not externally licensed
Key points
- Match the chart type to what the data actually represents: a line chart suits change over time, a bar chart suits comparing discrete categories, and a pie chart suits parts of a whole; using a pie chart for a time progression or a scatterplot for a two-option vote split forces a reader to reconstruct a relationship the chart’s own shape should have made obvious. Worked gallery of six chart types (bar, line, pie, scatterplot, histogram, treemap), each paired with the kind of comparison it’s suited for.
- Color should be chosen intentionally, with accessibility in mind, and — a data-visualization-specific refinement — colors within one visualization should generally differ in value and saturation, not hue alone.
- Test a color choice by converting it to greyscale: colors that stay distinguishable with hue removed are likely to work in full color; colors that collapse into the same gray were relying on hue alone rather than a genuine lightness difference.
- Differentiate data with variables beyond color: size, orientation, visual texture, shape, and position can each carry the same distinguishing role color does, and combining several gives a reader redundant cues instead of one single channel.
- Plan before adding detail: balance a visualization’s level of detail against legibility — too little context leaves a chart unclear, too much turns it into noise — and treat planning the visualization’s purpose as itself half the work, before laying out its specific marks.