Data Visualization 102: Common Mistakes When Visualizing Data


Data Visualization 102: Common Mistakes When Visualizing Data

There’s no doubt that data visualization is a powerful tool for getting information across to your readers.

However, if you visualize something poorly, your readers can be confused at best, or at worst, suspect that you are trying to intentionally manipulate them.

You can avoid these issues by watching out for these six common data visualization mistakes.

This is a follow-up post to an earlier article I wrote about how to visualize some of the most common simple types of data.

1. Mislabeling

Ever looked at a chart and noticed something was just plain wrong?


In the example above, two of the slices have their labels swapped. But as a viewer, this immediately throws me off kilter. Is it just the numbers that are swapped, or the entire label? There are situations where this information could be very important but I have no idea how to interpret it.

Here’s another example, from an actual textbook:



The axis looks just fine at a quick glance, but look at it a while longer and you’ll start to see something very odd going on…

Mislabeling your information is simply not an excusable mistake when it comes to visualizing data. No matter how you slice it, a mislabeled pie chart or wonky axis can completely obliterate the trust someone has in your information the moment they notice it.

Mislabeled data is easy enough to catch; just make sure you build enough time into the production process to get an extra set of eyes (or several!) on your visualization before it goes out to the public.

Specifically, instruct your test group to look for errors and mislabeled information to increase the chance that they will find it.

This is one of the worst mistakes to make and the easiest to fix – you just need to be sure to catch it!

2. Not to Scale

Scale is how we quickly look at a visual piece of information and draw conclusions based on what we are looking at. But when a chart or graph is not to scale, it is no longer possible to quickly gauge what it is telling you.

For example, a viewer looking at this chart might incorrectly assume that the majority of people use one of the top three US banking chains, instead of the other way around:


Without being to scale, the information is inaccurate and can at times seem untrustworthy, like in the example below:


This visualization from a Cyprus radio station shows a bar chart with the largest value dramatically highlighted.

However, upon closer inspection, it is just a tenth of a percent greater than the bar beside it! This potential for exaggeration is why viewers don’t trust charts that are not to scale.

3. Too Much Data

The goal of data visualization is to break down extensive data and complex topics in order to make them easier to understand, in a visual format.

One situation where this doesn’t work, though, is if you are trying to put too much data into a single chart.

A chart should have an overarching goal behind it. It should slice the data in one particular way in order to help the viewer draw a particular insight.

A chart should not try to pack in every single piece of information you have available. If it does, it will end up looking something like this:


The only insight this is giving me is a headache.

4. Impossible Comparisons

One of the great things about data visualization is that it makes data points easier to compare.

However, if you present the data in a series of separate charts, rather than highlighting the comparison you are trying to make in a single chart, the data quickly becomes impossible to compare in any meaningful way.


This is just three pie charts, but already it is starting to get difficult to draw comparisons between the percentages for the different years.

Imagine if this was six or even ten different pie charts. Very quickly it would become a nightmare to draw out any kind of trend from all those separate charts.

Instead, if you are trying to show a trend over the years, it is better to use another kind of visualization.

For example, a line graph with the decades on the X-axis and percentages up the Y-axis would be a much more compact way to show changing percentages over the years.

5. Data Viz “Just Because”

Each visualization you create needs to have a purpose. You can’t just stick a chart onto something and call it a data visualization.


In the example above, the creator of the infographic had some percentages, so they stuck a pie chart beside them for good measure.

The percentages do not correspond with the pie chart in any way, though – it isn’t even labeled!

Think about the purpose of the visualization you are using the data to create. If you can’t decide on a clear purpose, whatever you add will just be there to take up space.

6. Valuing Form Over Substance

Since data visualization exists to break down complex data to make it easier to understand, clarity is the ultimate priority.

However, sometimes our desire to fit the data to a particular theme or graphic seems to supersede this legibility, causing confusion.

Pictograms are a major offender here.

Pictograms, icons that stand in for larger numbers, are sometimes shown sliced in half or cut up in some way that makes them hard to recognize for what they are.

Similarly, some visualizations don’t use the space available very well.

In the example below, the viewer has to follow lines to the labels that indicate what slices of the pie chart represent – sometimes having to scroll up and down to view different parts of the same chart.

A more compact design would have been better here, where less scrolling was necessary to understand the graphic.


As nice as you want the visualization to look, make sure you stop to ask yourself if it will be easy to understand or read before publishing it.

Wrapping Up

When you create an infographic or any other kind of data visualization, you have to keep the viewer in mind.

You’re creating it to simplify data for that end user, after all, so you need to be sure that it actually does its job!

The data visualization mistakes above are some of the most common ones you see, and avoiding them is one of the fastest ways to ensure that your visualization is trustworthy and comprehensible.

Here’s a quick checklist for the mistakes to watch out for:

  • Avoid mislabeling your data.
  • Be sure to stick to a scale.
  • Don’t try to cram too much into one graphic.
  • Make it easy to compare and view trends across your data.
  • Every chart needs to have a clear purpose.
  • Don’t allow design to supersede substance.

Finally, you can always turn to a professional graphic designer for help. Our unlimited graphic design service is an affordable and headache-free solution that will sort out your visualization data in no time!

Got another mistake to add? Share it in the comments below!