1. Home
2. Guides

Guides

The parts of chart design that apply to every chart type: color, text, accessibility, layout and narrative. The second group covers one data source in depth, X (Twitter) analytics.

Fundamentals

• How to choose the right chart

  Pick a chart by the question it has to answer: comparison, change over time, distribution, relationship, part-to-whole or ranking. Includes a decision table.

• Using color in charts

  How to choose categorical, sequential and diverging palettes, use one highlight color, keep color meanings consistent, and design for color vision deficiency.

• Labels, titles and annotations

  Write chart titles that state the takeaway, label data directly instead of using legends, format axes and numbers clearly, and add notes and source lines.

• Making charts accessible

  Make charts readable for everyone: enough contrast, cues beyond color, alt text that states the finding, a data table, readable text, SVG for screen readers.

• Data storytelling: from numbers to a point

  Turn data into a clear point: find the one finding, add context and comparison, structure it as setup, finding and so-what, cut clutter, and frame it honestly.

Visualizing X (Twitter) data

• Charting X follower growth

  How to chart X follower growth from your analytics export: cumulative lines vs daily bars, follows vs unfollows, spikes, rolling averages and fair comparisons.

• Engagement heatmaps for X posts

  Build a day-of-week by hour heatmap for X posts: engagement rate vs raw counts, time zones, minimum posts per cell and how to read best time to post honestly.

• Network graphs of replies, reposts and mentions

  How to build a readable network graph from X replies, reposts and mentions: what nodes and edges mean, sizing, communities, and how to untangle a hairball.

• Comparing tweet performance

  Compare X post performance fairly: impressions vs engagement rate, dot plots and slope charts, sorting, the viral outlier problem, medians and small multiples.