Start hereWhat makes a chart explain rather than decorate?
A chart is an encoding of evidence. Good visualisation chooses position, length, colour and annotation to make the intended comparison effortless and honest.
Building interactive view…
Technical lensFormalise what the visual is doing
Chart selection follows the analytical task: comparison, trend, distribution, relationship, composition or uncertainty. Axis scales and aggregation choices can distort perception.
Technical questionUse a tiny case to make the mechanism observable. Chart selection follows the analytical task: comparison, trend, distribution, relationship, composition or uncertainty. Axis scales and aggregation choices can distort perception. Verify one intermediate quantity, state change or mapping independently; then predict the consequence of this change: Truncate an axis or switch aggregation and inspect how the apparent story changes.
Practitioner lensUse it responsibly
Write the takeaway first, then choose the simplest encoding that supports it. Label units, show uncertainty when relevant and remove non-informative decoration.
Transfer test3D effects that distort perceived magnitude.
Worked explorationUse the visual as an experiment, not decoration
Show the same category values in a correctly based bar chart and in a truncated-axis bar chart. Compare how the visual impression changes even though the numbers do not. Add direct labels and remove unnecessary decoration.
Technical lens
Chart selection follows the analytical task: comparison, trend, distribution, relationship, composition or uncertainty. Axis scales and aggregation choices can distort perception.
Practitioner check
Write the takeaway first, then choose the simplest encoding that supports it. Label units, show uncertainty when relevant and remove non-informative decoration.
Prediction before interactionTruncate an axis or switch aggregation and inspect how the apparent story changes.
Exploration walkthroughTurn the interaction into an evidence trail
Show the same category values in a correctly based bar chart and in a truncated-axis bar chart. Compare how the visual impression changes even though the numbers do not. Add direct labels and remove unnecessary decoration. Before moving the control, state your prediction. After the visual changes, name the specific state, statistic, boundary or mapping that changed and explain why that change is consistent—or inconsistent—with your prediction.
- Record one observable quantity before the interaction and the same quantity afterwards.
- Change one factor at a time so the causal effect of the control is inspectable.
- Use an edge or failure case to discover where the concept stops behaving as the simple story suggests.
