"Revenue up 200%!" — accompanied by a dramatically rising line chart. But look closely at the Y-axis: it starts at 98 and ends at 100. That "explosive growth" is actually a change from 98 to 100. Charts are powerful tools for communicating data, but they're also one of the most easily misused media for manipulating perception.
1. The Most Common Chart Deception: Truncating the Y-Axis
The truncated Y-axis is probably the most widely used visual deception technique in charts. The trick is to start the Y-axis at some number above zero, making small differences look enormous.
Consider this scenario:
- Option A: 78% satisfaction
- Option B: 82% satisfaction
With a Y-axis starting at 0, the two bars are nearly the same height — because they differ by only about 5%. But with a Y-axis starting at 75%, Option B's bar will be several times taller than Option A's — conveying an impression of overwhelming superiority for what's actually a 4-percentage-point difference.
When is truncating the Y-axis legitimate?
Not all truncation is deceptive. If values are inherently concentrated (like a stock price moving from $499 to $503), starting from zero makes all the variation disappear — which is bad design in the other direction. The key is clear labeling: a truncated Y-axis must use a visible break symbol (zigzag or double-slash) to indicate that zero is not the starting point.
2. 3D Charts: Stylish but Systematically Misleading
3D effects make charts look polished and impressive. But 3D introduces visual distortion in almost every case:
The problem with 3D pie charts
The moment you add 3D to a pie chart, slices closest to the "viewer's perspective" appear visually larger than those in the background — even when both sectors represent equal values. This is a natural consequence of perspective, but in data visualization it's a serious distortion. It's also commonly exploited: place the number you want to emphasize up front, add 3D, and it looks bigger than it is.
The problem with 3D bar charts
3D bar charts make it difficult to judge the true tops of bars, because the viewing angle makes bars at different depths appear at different heights. The ability to read precise values drops significantly.
Rule of thumb: Unless the chart is purely decorative, don't use 3D effects.
3. Pie Charts: Rules for Correct Use
Pie charts are among the most misused chart types. A pie chart is only appropriate when:
- The values add up to exactly 100% (part-to-whole relationship)
- There are 5 or fewer slices — more than that, and small slices become impossible to compare
- You're not trying to compare the same category across two different time points or groups (use grouped bar charts for that)
Common pie chart design traps
| Design Problem | Effect | Correct Approach |
|---|---|---|
| "Exploded" slice to emphasize one segment | Makes that segment visually prominent and appear larger | Avoid explosion effects except for labeling purposes |
| 3D perspective | Front slices appear visually enlarged | Always use 2D |
| Illogical slice ordering | Eyes jump around, comparisons are difficult | Sort by value, or use a logical order (e.g., clockwise largest-first) |
| More than 7 slices | Small slices become nearly indistinguishable | Group small items into an "Other" category |
4. Color Is a Manipulation Tool Too
Chart color design looks like pure aesthetics, but color has a direct effect on cognitive judgment:
Unbalanced contrast
Using vivid red for your product and gray for the competitor — even if the numbers are similar, viewers' eyes go straight to the vibrant color, implicitly suggesting that your product matters more.
Colors carry implicit emotional valence
Red is typically interpreted as danger or negative; green as safe or positive. A chart that colors the competitor's numbers red and yours green has already completed the "good/bad" judgment at the visual level, without a single word of explanation.
Gradient colors mislead about continuity
In heat maps and choropleth maps: human eyes perceive brightness differently across hues, so the visually "brightest" area doesn't necessarily correspond to the highest value. Perceptually uniform color scales like Viridis are more scientifically sound.
5. A Checklist for Reading Charts Without Being Fooled
Next time you encounter a chart, run through these checks:
- Check the Y-axis starting point: Does it start at 0? If not, is the difference really as large as it looks?
- Check the sample size: How many people or observations is this based on? Is it labeled?
- Check if the chart type is appropriate: Do the pie chart values add up to 100%? Is there a 3D effect?
- Check the color design: Is one side getting a visual boost from color choices?
- Check the actual numbers: How large is the gap between the visual impression and the raw data?
- Ask: "Who made this data, and what's their incentive?"
6. Integrity Principles for Making Your Own Charts
- Start the Y-axis at 0 unless there's a compelling reason (and clearly mark the break)
- No 3D effects unless the chart is purely decorative
- Always label sample size and data source
- Make sure the chart's "feeling" matches the actual numeric difference — if the visual impression is dramatically larger than the real gap, ask yourself why
- Respect your readers: give them the complete picture and let them draw conclusions, rather than engineering the conclusion you want them to reach
Summary
- Truncated Y-axes are the most common visual deception — small differences can be amplified into seemingly huge ones
- 3D effects cause visual distortion in nearly all data charts and should be avoided
- Pie charts have strict usage requirements; too many slices or 3D treatment both mislead
- Color choices influence readers' judgment at the subconscious level
- When reading any chart, always check the Y-axis starting point, sample size, and data source first