Good chart design begins with the question a reader needs to answer, the decision they need to make, and the evidence required to support it. The chart comes later.
A chart can be accurate and attractive yet still be useless if it answers the wrong question. It can contain the right numbers and remain difficult to read if it asks the eye to compare angles, areas or colours when position and length would communicate more clearly. It can be immediately legible and still mislead if the denominator, benchmark or uncertainty is missing.
Chart design is therefore not a matter of taste. It is a decision-support discipline grounded in evidence, visual perception and the context in which information will be used.
Three parts, one method
The guide separates the underlying discipline from the practical work of selecting and styling charts.
Part I — Foundations of Effective Chart Design
Start with purpose and evidence, then design for perception, build a clear information hierarchy and test whether readers can make the intended judgement. This part explains why effective chart design is learned rather than improvised from preference.
Part II — Choosing the Right Chart
Assume the business question, decision and evidence have been established. Move from analytical intent to data shape, then choose the visual form that makes the answer easiest to see. The interactive taxonomy above visualises that path.
Part III — Colour and Data Palettes
Treat colour as an analytical encoding rather than a finishing treatment. Choose among sequential, diverging, categorical, cyclic and semantic palettes, then test contrast, perceptual separation and accessibility in the chart where the palette will be read.
How the parts work together
The three parts are sequential without being restrictive. If a chart request begins with a visual preference, return to Part I and clarify its purpose. If the purpose and evidence are sound but the form is uncertain, begin with Part II. If the chart works structurally but colour is carrying too much meaning or too little, use Part III.
The method is deliberately recursive. Testing may reveal that a chart answers the wrong question. A chart-selection problem may expose missing evidence. A palette that works as swatches may fail once applied to thin lines, small marks or a dense dashboard. Good design moves backwards when the result demands it.
The Chart Index
The Chart Index is the detailed reference layer beneath this guide. Every major chart form receives its own article: what it is, when to use it, when not to use it, the design principles that make it effective, its anatomy and its related forms.
Use the field guide to establish the reasoning. Use the Index once the candidate forms are known and the execution details matter.
Beyond a single chart
The same discipline applies to dashboards, but a dashboard is more than a collection of individually sound charts. It introduces information architecture, monitoring, interaction, navigation and coordination across views. Those concerns deserve their own field guide rather than a closing section added to chart selection.
The test remains the same at every scale: can the reader see the evidence, reach a sound judgement and act with greater confidence?