Data-ink encodes values as the share of ink that represents data versus decoration. Edward Tufte, The Visual Display of Quantitative Information, 1983.
In reports, ads and assignments it fits auditing slides, dashboards and science figures. Skip it when a poster that must shout from three metres — ads may keep some junk on purpose.
Principles
- Encoding — Data-ink encodes the share of ink that represents data versus decoration.
- History — Edward Tufte, The Visual Display of Quantitative Information, 1983
- Fit — auditing slides, dashboards and science figures
- Shared baseline — The eye compares length and position more accurately than area.
- Audience — PowerPoint, Excel or Google Sheets and Meta, Google and retail print share the same geometry.
Misconceptions
- Pretty equals true — Decoration rarely persuades; it slows reading.
- 3-D helps — Perspective distortion wrecks comparison. Keep it flat.
- More series is richer — Do not stack stories. One Data-ink, one claim.
- Default palette is fine — Rainbow defaults fail color-blind viewers and grey print. Change Excel or Google Sheets defaults.
Why it matters
- Work — Work: auditing slides, dashboards and science figures. Readable in four minutes in PowerPoint.
- Ads — Ads: one number and one shape speak first in Meta, Google and retail print.
- School — School: for lab reports and poster boards, pick the form that answers the rubric.
- Craft — Craft: honour the share of ink that represents data versus decoration and decoration falls away.
Data-ink: honour the share of ink that represents data versus decoration, and switch when a poster that must shout from three metres — ads may keep some junk on purpose.