The fastest way to lose a stakeholder's trust in an otherwise solid analysis is a chart that makes them work too hard, or worse, a chart that quietly misleads them. None of these five mistakes require advanced design skill to avoid — they just require noticing them.
Pie charts are genuinely useful for showing a simple part-to-whole relationship with a handful of categories. Past four or five slices, humans lose the ability to compare angles accurately, and the chart becomes decorative rather than informative. The fix: switch to a horizontal bar chart, sorted by value. It communicates the same comparison faster and scales to far more categories without becoming unreadable.
Starting a bar chart's y-axis at 90 instead of 0 can make a 2% change look like a dramatic swing. Sometimes this is genuinely useful for showing small but meaningful variation — but done without a clear label or annotation, it reads as manipulation, whether or not that was the intent. The fix: default to a zero-based axis for bar charts. If you have a real reason to truncate (showing subtle variation in a stable metric), label the axis clearly and consider adding a note explaining why.
Assigning a different bright color to every category "because it looks nice" adds visual noise without adding information, and it actively works against faster comprehension. The fix: use color deliberately. One accent color to highlight the single most important data point or category, with everything else in neutral gray, does more communicative work than ten distinct colors competing for attention.
A table showing "$47,382.19" when the audience only needs to know it's "about $47K" forces extra cognitive work for no added value, and it can make a rough estimate look falsely precise. The fix: round to the level of precision the decision actually requires. Executives comparing regional performance rarely need the exact cent — round numbers speed up comprehension without losing anything that matters to the decision at hand.
A chart titled "Revenue by Region" tells the reader what the chart contains, not what it means. Readers are left to derive the insight themselves, and different people often walk away with different conclusions from the exact same chart. The fix: write titles as the finding, not the topic — "Revenue Grew Fastest in the Southeast Region" tells the reader the takeaway immediately, and the chart becomes the supporting evidence rather than a puzzle they have to solve.
Every mistake on this list shares the same root cause: designing the chart around what's easy to produce rather than what's easy for the reader to understand. Fixing all five doesn't require design training — it requires pausing before you ship a chart and asking, "if someone looked at this for five seconds, would they get the right idea?"
Are pie charts always wrong to use? No — for two or three categories with a clear part-to-whole story, they work fine. The problem is specifically overuse past the point where angle comparison stays accurate.
Is it ever okay to truncate a y-axis? Yes, when clearly labeled and there's a real reason (showing subtle but meaningful variation in an otherwise stable metric). The problem is doing it silently, not doing it at all.
How do I pick which single color to highlight in a chart? Highlight whatever the chart's takeaway is actually about — the region that grew fastest, the product line that's underperforming — and leave everything else in neutral gray so the highlight actually stands out.
Where can I learn more about this systematically? See our Data Communication Concepts course for a structured, tool-agnostic path through these principles, or our guide on presenting data to executives for the narrative side of the same problem.
These five mistakes are common precisely because they're easy to make without noticing — none require bad intentions, just a lack of a final review pass. Building the habit of checking for them before you share a chart is a fast, low-effort way to meaningfully raise the credibility of everything you present.
