10 Common Data Visualization Mistakes (and How to Fix Them)
Updated · 2 min read
Most misleading charts aren’t made on purpose. They come from defaults and habits. Here are the mistakes we see most, and how our tools help you avoid them.
1. Truncated bar chart axes
Starting a bar axis above zero exaggerates differences. Bars must start at zero; our bar graph maker enforces it.
2. 3D effects
Perspective makes near elements look larger. Use flat charts. Our pie chart’s optional shading is decorative and never tilts the geometry.
3. Too many slices or series
Beyond six or seven categories, colors become hard to tell apart. Group small items into “Other”, or split the chart into small multiples.
4. Rainbow color scales
Rainbow scales create false boundaries. Use single-hue or viridis-style scales for ordered values, and diverging scales only around a meaningful midpoint.
5. Relying on color alone
Around 8% of men and 0.5% of women have some form of color-vision deficiency. Add direct labels, distinct marker shapes or pattern fills.
6. Pie charts for non-parts
If values don’t sum to a whole, for example multi-select survey answers or averages, a pie chart is wrong. Use bars.
7. Misleading dual axes
Two independently scaled axes can make any two series look correlated. Label both axes and avoid implying causation.
8. Area scaled by diameter
Doubling a circle’s diameter quadruples its area. Bubble and polar-area charts should scale by area, as ours do by default.
9. Averages without spread
Two groups with the same mean can be completely different. Show the distribution with a histogram or box plot.
10. Missing context
A chart without a title, units, source or sample size forces readers to guess. Use the title for the takeaway (“Sales doubled in Q4”) and the subtitle for units and the base.