Things to avoid in Data Visualisation

I have been preparing tons of visualizations, asking investigative question looking at billions of data every week, I believe following are the common data visualization mistakes:

Pies & donuts with too many segments:

Comparing angles are not something humans are excellent in. If you got too many segments, in that case, it is better to use bar or column charts instead.
 

Line charts for categorical comparisons:

Line charts should be used to show trends. Using them for categorical data (vs. time-series) is misleading, and can suggest relationships or patterns that don't exist.
 

Treemaps for non-hierarchical data:

Treemaps are designed to show hierarchies (like subcategories within categories). For simple categorical comparisons, use a bar or column chart instead.
 

Unsorted data:

Don't expect viewers to make their own visual comparisons. Use intuitive sorting rules to organize your data and tell a clear story.

Too much noise, too little focus:

Dashboards should clearly communicate the most important information, and often the best way to do this is by eliminating noise. While it's tempting to add that cool ribbon chart or 3-D effect, remember that CLARITY and INSIGHT are far more important than variety or visual appeal.