A bad chart is worse than no chart. It misleads, confuses, and undermines your credibility. The good news is that most charting mistakes are simple, predictable, and easy to fix once you know what to look for.

Here are the seven most common ones.

1. Truncated y-axes

One of the most misleading tricks in data visualization is starting the y-axis at something other than zero. This makes small differences look enormous.

A bar chart showing revenue of 98, 99, 100, and 101 looks like a flat line when the y-axis runs from 0 to 150. But if the axis starts at 97, those same bars look like a dramatic roller coaster.

Fix: For bar charts, start the y-axis at zero. For line charts, truncation is more acceptable when the absolute scale would compress all variation into a flat line — but label the axis clearly so the reader understands the scale.

2. Pie charts with too many segments

Pie charts are designed for two to five segments. Beyond that, the slices become too thin to read, the legend becomes a chore, and nobody can reliably compare angles. A 12-segment pie chart is always wrong.

Fix: If you have more than five categories, group the smallest ones into an "Other" segment, or switch to a horizontal bar chart. Bars are almost always easier to compare than slices.

3. 3D charts

Three-dimensional charts add visual complexity without adding information. The depth creates perspective distortion that makes values near the back of the chart look smaller than equivalent values at the front. They also make it difficult to see the reference line (the baseline).

Fix: Use flat, 2D charts. Every time. There are no exceptions.

4. Missing or misleading chart titles

A chart titled "Revenue" tells the reader almost nothing. Revenue where? Which time period? Compared to what? In what currency?

The chart title is the most important piece of text in the entire visualization. It should answer the "so what?" question — the insight the chart is intended to communicate — not just describe what is plotted.

Fix: Write the title as a declarative sentence: "European Revenue Grew 23% in Q2 2025" rather than "Q2 Revenue by Region." The sentence-title format forces you to think about what you actually learned from the chart.

5. Dual y-axes

Putting two different scales on a single chart — one on the left, one on the right — creates the illusion of a relationship between two series that may or may not actually exist. By scaling each axis independently, you can make any two trends appear to correlate.

Fix: Use separate charts for series with different units. If you must compare two series with different scales, show them separately but aligned vertically so the reader can compare them without the scales interfering.

6. Rainbow colour schemes

Using a different colour for every category looks festive but rarely communicates anything. When every bar is a different colour, the eye has to work hard to decode the legend on every single data point. Colour becomes noise rather than signal.

Fix: Use a neutral base colour (grey or light blue) for all categories, and highlight only the category you want to draw attention to in a stronger colour. If you're comparing groups, use a consistent palette where each group always gets the same colour across all charts in a report.

7. Showing averages without showing spread

Average scores, average revenue, average temperatures — averages hide enormous amounts of useful information. Two datasets can have exactly the same mean while having completely different distributions.

A chart showing that average customer satisfaction is 7.2/10 tells you very little. Are most customers giving 7s and 7.5s? Or is half giving 3s and half giving 10s? Those are completely different situations requiring completely different responses.

Fix: Pair averages with a measure of spread — a standard deviation label, error bars, or, better yet, replace the bar chart entirely with a box plot or histogram that shows the full distribution.

The underlying principle

Most visualization mistakes share a root cause: optimising for how the chart looks rather than what it communicates. A chart that impresses in a presentation but misleads the audience has failed at its actual job.

Design for clarity. Remove everything that doesn't help the reader understand the data faster or more accurately. Then double-check that what remains is honest about what the data actually shows.