How to Choose the Best Charts for Your Infographic

By Midori Nediger, Jun 29, 2018

best chart

The best infographics fuse data with text and visuals to tell a persuasive story.

Unfortunately, you can’t just plop some numbers into your infographic and expect your readers to gain anything from it. You have to put a some thought into the presentation of the data to create clarity and boost understanding.

But I guarantee you, that little bit of extra effort you put into choosing the best chart for your data will be worth it! When used appropriately, charts can:

So how do you go about choosing the best charts for your infographic?

best chart

 

Use the ICCOR method to choose the best chart for your infographic

At Venngage, we’ve developed the ICCOR method to help you choose the best charts for your infographic.

There are two steps to making the ICCOR method work for you:

  1. Determine your visualization goal
  2. Choose the best chart to achieve that goal

Step 1: Determine your visualization goal

If you try to dive into visualizing your data without any clear goals, you’re never going to produce anything useful.

Garbage in, garbage out, right?

Instead, start by defining a single visualization goal for your chart. That goal should line up with one of our ICCOR categories, either to:

    • Inform: convey a single important message or data point that doesn’t require much context to understand
    • Compare: show similarities or differences among values or parts of a whole
    • Show Change: visualize trends over time or space
    • Organize: show groups, patterns, rank or order
    • Reveal Relationships: show correlations among variables or values

best chart

 

Not sure what your goal should be? Spend some time getting to know your data. Look for patterns, groupings, trends, outliers, and key data points. The things that stand out to you are the same things that you should emphasize to your readers.

Step 2: Choose the best chart to achieve that goal

With your goal in mind, it’s time to choose your chart!

There are a number of charts that can work for each ICCOR goal.

Let’s review each one with some chart design best practices for each.

Best charts for informing

So you have a single data point or key message that you want to hit your readers over the head with.

The easiest way to do that in an infographic is to simply use big, bold (colorful) text.

best chart

Source

The nice thing about this strategy is it gets right to the point–there’s no room for misinterpretation.

The drawback?

Sometimes a single number, without any context, might not seem very significant.

This problem is easily remedied. Use simple symbols or icons to add a bit more context–like an arrow to emphasize growth or decline:

best chart

 

Or an icon chart (an icon with two fill colors, where the ratio is represented by the height of the main color) to make proportions more tangible.

best chart

 

A more accurate way of representing a proportion is by directly spelling it out in a pictogram, a matrix of icons in which one icon represents a unit or group of units. When making pictograms, try to stick with nice round numbers like 10 or 100 that are easy to grasp:

best chart

 

Best charts for comparing

If you want to show similarities or differences among values or parts of a whole, you’ve got a lot of options.

Use a bar or column chart to compare independent values

We, as readers, are particularly good at comparing the length of bars in a bar chart (in contrast to the segments of a pie chart, for example), making bar and column charts the best charts for showing comparisons.

For the most part, bar charts and column charts can be used interchangeably. However, if you want to compare the change in some metric over time, it’s better to stick with a column chart so that time is represented from left to right.

best chart

 

If you happen to have long labels for each data point (like when you’re charting survey results, for example), the horizontal bar chart is a better choice. It allows extra space for each label so that hard to read vertical or angled text can be avoided:

best chart

 

Note the order of the bars here–since the dataset has no inherent order, sorting the bars from greatest to least adds an extra dimension of information. Also note the ratio of bar width to bar spacing. The space between the bars should be about half of the bar width for optimal readability.

While bar and column charts are the best choice in terms of data readability, infographics sometimes call for something a little more unique and engaging.

Enter: the bubble chart.

Use a bubble chart to compare independent values with clear outliers

In a bubble chart, values are represented by circle area instead of bar length. Since we’re not particularly good at making precise judgements of circle area, it’s only a good idea to use bubble charts when there are clear outliers or large gaps between each value. I’d also recommend explicitly labeling each bubble with its value.

best chart

 

Use a pie chart to compare parts of a whole (composition)

The simplest way to visualize simple part-to-whole relationships is with a pie chart. Make your pie charts easy to read by:

  • Ordering segments from greatest to least
  • Starting the first segment at 12:00 and continuing clockwise
  • Limiting the chart to a maximum of 7 segments

best chart

 

While pie charts are ubiquitous and easily understood, they do have their limitations (i.e. when comparing the composition of multiple values). Instead:

Use a stacked bar or stacked column chart to compare the compositions of multiple values

best chart

 

Even stacked bar charts are fairly limited in their uses, though. They only really work when you have very few bars and very few segments, or when differences between bars are very large. If you’ve got lots of data or the differences within your data are subtle, a series of bar charts would be a better choice.

best chart

Source

Basically–bar charts are the most reliable method for visualizing comparisons.

Best charts for showing change

There’s a little more flexibility when it comes to visualizing trends over time or space.

Use a line chart or an area chart to show changes that are continuous over time.

Line charts are the most effective chart for displaying time series data. They can handle a ton of data points and multiple data series, and everyone knows how to read them. Just make sure your points are ordered such that time runs from left to right, and use consistent time intervals.

best chart

 

Area charts can be more aesthetically pleasing, but require some more finesse. They’re only really appropriate to use with less than four data series, and it’s a good idea to introduce some transparency to make sure each area is visible.  

best chart

 

But time series data isn’t always purely numerical!

Use a timeline infographic to visualize events over time

best chart

 

Use a choropleth map to show location data.

A chloropleth map is a map crossed with a heatmap. They’re usually shaded in proportion to the value of some variable, and can be used to show location-based patterns:

best chart

 

Use a map series to show changes in location data over time

best chart

 

Best charts for organizing

Charts for organizing data or information in an infographic can take many forms, depending on whether you want to show groups, patterns, rank, or order. I’ll just touch on a few essentials.

The simplest form of infographic organization is a list.

Use a numbered list to show rank or order when you want to provide textual information about each element (for example, when describing a linear process)

best chart

 

Use a table to organize when you want readers to be able to look up specific values

Tables are best when precise values are required, when viewers need access to individual numbers, or when the dataset contains a number of different units (which would be challenging to visualize on a single graph).

best chart

 

While tables can provide a lot of detail, they don’t give you a sense of shape of the data, so it’s often a good idea to include a tables as a supplement to a more visual chart.

Use simple boxes, borders, arrows, and lines to visually organize groups (as seen in Venn diagrams, mind maps, and flow charts)

Most of the other methods of organizing information rely on two basic principles:

  • Using shapes and borders to create groups
  • Using lines and arrows indicate the relationship between groups (i.e. connect groups)

Some popular infographic charts that use these principles are the Venn diagram:

best chart

 

And the mind map:

best chart

 

To find the best organizational chart for your infographic, think about the structure of your information and find a chart that reflects that structure.

Best charts for revealing relationships

Two key types of relationships you might want to reveal within a dataset are correlation and distribution.

Use a scatter plot to reveal the correlation and distribution of a two-variable dataset

Scatter plots are the easiest way to explore a potential correlation between two variables, and can also reveal the distribution, clustering, and outliers of a dataset.

best chart

 

Scatter plots are handy because they don’t require the data to be aggregated or pre-processed to reveal the distribution of a dataset.

If you’re willing to do some preprocessing or you already have nicely aggregated data:

Use a histogram to reveal the distribution of a single variable

Histograms are the standard way to display age demographic data, and are great for all kinds of distributions.

best chart

 

Quick tips for infographic chart design

Now that we’ve covered how to choose the best chart for each ICCOR goal, it’s time to talk chart styling.

At best, properly styled charts can make your infographics clearer and more persuasive.

At worst, poorly designed charts can be misleading, create confusion, undermine the credibility of the author. If you’re interested, check out how bad chart design was used in the media to skew data during the 2016 U.S. election.

To make sure you create charts that are clear, accessible, and 100% above board, let’s review some chart design best practices.

Start chart axes at zero, and use consistent intervals for tick marks

One of the biggest data visualization faux pas is axis truncation. Starting the axis of a bar or line chart at some value other than zero is a sure-fire way to distort your data:

In most cases, it’s best to start your axes at zero. If you choose not to, make sure you have a  good reason for it. Similarly, make sure the scale of your chart is apparent by labeling your axes with regularly-spaced tick marks at logical intervals (such as every 5, 10, or 100 units).

Reduce non-data ink

Your data should always be the star of the show.

Data visualization guru Edward Tufte coined the term “data-ink ratio” to emphasize the importance of minimalism in chart design. The general idea: a large share of the ink required to print a graph should directly represent data-information

Basically, things like borders, gridlines, background colors, and other extra decorations should take a backseat to the points, bars, or lines that actually represent the data.

Do your best to lighten labels and gridlines, diminish colors, eliminate backgrounds, avoid borders and outlines, and above all…get rid of any 3D or other “special” effects.

best chart

GIF Source

Don’t forget labels, legends, and annotations

Labels are your friends.

Chart labels, headings, and annotations should be used to explicitly state the things you want your readers to take away from your chart, to eliminate the possibility of misinterpretation.

Don’t be afraid to be redundant, either. It’s okay to repeat key takeaways in both a subheading and an annotation.

best chart

 

When labeling lines, bars, or points, try to avoid relying on a separate legend. Place labels as close to their corresponding data points as possible, so that your reader doesn’t have to search to match up labels with data points.

Oh, and don’t forget to include your data sources. Link to your data source at the bottom of your infographic in small type (less than 10 pt size).

Use color for function, not decoration

Color use plays a significant role in how “busy” your infographic will appear.

To make sure things don’t get too messy, don’t use more than six colors in any given chart. When you’ve got more than one chart in an infographic, use those colors consistently across all of your charts.

best chart

 

For more complex chart charts, choose one key color to distinguish more important values from less important values (and use gray to attenuate the data that you want to be less of a focus).

best chart

 

In contrast, don’t rely on color to make your chart legible–your chart should be just as meaningful in black and white as it is in color.

best chart

Order data logically

Present your data in an order that your reader will expect.

Have a time series? Order it with the earliest data point on the left and the latest data point on the right.

Have a pie chart? Starting at twelve o’clock, order segments clockwise from largest to smallest.

best chart

 

Representing categories that have no natural order? Consider ranking them from largest to smallest.

best chart

 

Keep it simple!

Finally, when visualizing data in an infographic, I urge you to exercise restraint. Don’t overload your charts with too much data.

Infographic data must be easy to understand at a glance. If there’s too much going on in the chart, your reader is likely to give up and move on.

Some rules of thumb:

  • Use less than 6 lines in a line chart
  • Use less than 10 bars in a bar chart
  • Use less than 7 segments in a pie chart

If this means manipulating your data (by removing points, grouping points, or by looking at shorter spans of time), take time to consider the tradeoff between readability and data accuracy. Try to reduce your dataset down to its most essential pieces without introducing inaccuracy or skewing your data.

If you can’t get around including a ton of data, it’s not the end of the world. Contrasting a single bright color with muted greys can help your readers make sense of even hundreds of data points:

best chart

Source

A good way to make sure that your chart is simple enough? Have a friend look it over, and resist the urge to explain it to them. If your friend can’t understand your chart without any verbal explanation, it’s too complex for an infographic.

Alternatively, do a squint test. Squint your eyes until you can’t read any text–do you still get something from the chart? If not, simplify your charts a little more.

best chart

Source

So to sum all of this up:

  • be diligent with your axes
  • use color thoughtfully
  • minimize the non-data ink
  • don’t forget about labels and annotations
  • cite your sources
  • order your data logically
  • don’t overload your chart with too much data

Luckily for you, the default styling of Venngage’s chart widgets take care of a lot of this chart design stuff for you. Just click the chart you want to use, pop in your data, and you’re good to go!

 

Conclusion

The power of well-presented data should not be underestimated, and the right chart can make all the difference. Use the ICCOR Method to choose the best charts for your infographic data.



About Midori Nediger

Midori spreads visual communication tricks and tips as an Information Designer at Venngage. She’s particularly interested in helping people communicate complex information. Connect with her on LinkedIn and on Twitter @MNediger.