Data storytelling is one of those “buzzwords” that in actuality is not really a buzzword–it’s reflective of a necessary change in the way humans are making sense of the inordinate amount of information out there today.
It is estimated that by 2020, the digital universe will reach 44 zettabytes, or 44 trillion gigabytes, which is 10 times as much information as there was in 2013.
If you can’t fathom how much that is, then imagine the amount of information represented by the memory in a stack of tablets stretching from the Earth to the moon–six and a half times. Now that’s big data!
It’s no wonder then that making sense of otherwise meaningless numbers through storytelling is a must-have skill for the future. We may live in a digital era, but storytelling is still our primary tool for making sense of the world around us.
Sadly, many communicators are scared to death of numbers. They would rather stick to being wordsmiths, using the occasional stat to back up their points. But in the era of open access data, in which even math-averse journalists are becoming adept at handling large volumes of data, you can only get by with this mindset for so long.
If you’re determined to venture into this brave, new world of data viz, let’s start with some of the basics of telling stories with data and common missteps you should avoid.
The Science Behind Good Data Visualizations
The Illusion of Human Perception
The first step to creating effective and persuasive data visualizations is to understand how human perception works.
Contrary to what you may have learned in high school, the process of perceiving an object is much more complex than the conceptual model of a digital video camera, in which our eyes act as a lens, our optic nerves as cables and our brains as processors and hard drives.
Alberto Cairo, in his book The Functional Art, delves deeper into the mechanics of human perception and explains how illusion plays an important role in the way we see the world.
For example, as you’re looking at the screen in front of you, it seems like you can see everything within an 180-degree angle, but in actuality, you can only see with full accuracy those things that lie in a very narrow 2-degree field straight ahead of you.
How is it then that we don’t see a blurry mass of things?
Thanks to rapid eye movements called saccades, our eyes quickly dart around scenes to create composite images from the aggregate information, thus creating the very believable illusion that our eyes act like a 180-degree lens.
What we think we see looks something like the image on the left below, but what our eyes are actually doing is sending “small snapshots” of different points, like the ones on the right.
What does this have to do with information design? It’s useful to know that our eyes don’t fixate on random points in a scene or image but rather prioritize. They first detect basic features and focus on things that stand out such as moving objects, bright-colored patches and uncommon shapes.
Preattentive Attributes
These basic features are also called preattentive features: Before we’re even consciously aware of them, our brain is already processing these properties.
For example, when our eyes look at a scene, such as the one above, the first thing our brains detect is the difference between the background and foreground. They can detect where the passageway starts and ends, where the trees begin and end, etc. The higher the contrast between the elements, the easier it is for the brain to discern the difference.
Because our brains are designed in this way, we can quickly make out the bear in the first two images below but not the last:
Since the brain is much better at discerning differences in color rather than shape, the best data visualizations deliberately use shade differences to draw attention to certain key pieces of information.
To save time, the brain has also evolved to group similar objects together and quickly identify objects that are different.
Take a look at these images below. You’ll see that without thinking about it, your brain has already recognized a pattern in each one and identified the contrasting element.
Without a doubt, this ability to preattentively detect features is one of the most important tools an information designer can use to create effective and persuasive data visualizations.
Now that we’ve looked at some of the science behind human perception, let’s take a dive into the actual process involved in telling stories with data:
Step 1: Resist The Urge to Immediately Choose a Chart
For most people, the process of creating charts and graphs seems almost intuitive. You select the rows and columns you want to visualize, then click on one of the Excel chart options and voila!–your chart is complete.
But if you want to go beyond simply presenting information to telling stories with data, then the process is a bit more involved.
The book Good Charts, by Harvard Business Review, suggests resisting the urge to immediately choose a chart by instead asking ourselves these two questions:
- Is the information data-driven or conceptual?
- Is the objective to make a declarative statement or explore something?
Once you’ve answered these, you can plot your response in one of the four quadrants below to get a better idea of what kind of visualization you should use.
If you’ve plotted your response in the top-left quadrant, then you will most likely need to simplify an idea using diagrams and other illustrations. Often, metaphors such as mountains and pyramids are used in these types of visualizations, as well as cultural conventions such as hierarchies and symbols like arrows and icons.
If you’ve plotted your response in the the bottom-left quadrant, then your goal is to explore a topic through conceptual visuals. This is most commonly used in whiteboard sessions where you want to, let’s say, plan a business process or diagram a system without using concrete data.
In contrast, the third category to the bottom right requires a visualization that is more complex than all the others. Here, you will most likely deal with multiple large data sets and work in conjunction with a data scientist to manage and present the data in the form of an interactive visualization to uncover patterns and trends.
Finally, the category in the top-right quadrant includes common visualizations such as bar charts, line charts and scatter plots. The goal here is to affirm or provide context, so the focus should be on defining a simple narrative and driving home a clear point through effective design decisions.
Since we could spend hours on each one of the visualization types, for the purposes of this post, we’ll only go into the process for creating declarative visualizations, which are found in the top half of the diagram above.
Step 2: Consult For Context Before You Start
Another commonly overlooked step is taking a bit of time to think about the context of your visualization.
For example, do you know who your audience is? The better you know your audience, what their needs and challenges are, then the more likely you will be able to approach a topic in a way that resonates with them.
It’s also important to ask yourself: What do you want your audience to know or do? Do you want them to make a certain decision? Or simply start a discussion on a topic?
Also, what setting will this be used in? Is it part of a live presentation, where you’ll have more control over the way information is presented? Or a printed document, where the reader decides how fast and how deep he or she will delve into the information?
All of these responses will be important in answering the last question: how can you use your data to make this point? This brings us to the next point…
Step 3: Define The Focus of The Graphic and The Story You Want To Tell
The key to all effective communication, whether it be through writing, information visualization or graphic design, is to hone in on your main message.
In Storytelling With Data, Cole Nussbaumer Knaflic discusses the need to boil down your message to a “so what” statement. This is harder than it sounds, so to make it easier on you, articulate your story to a friend or colleague who is unfamiliar with the data in three minutes or less. Next, take this a step further by condensing it into a single sentence.
According to Nancy Duarte, author of Resonate, this sentence should have three components:
- It must state your unique point of view.
- It must define what’s at stake (why your audience should care about it)
- It must be a complete sentence.
An example of a “so what” sentence: “Our software allows you to create visual content in less than half the time it takes to use traditional graphic design tools, thereby saving DIY designers time and resources.”
Step 4: Use Physical Markers, Pencils and Paper to Storyboard Your Idea
Before you’re tempted to open PowerPoint or some other desktop application, find a few colored markers and paper and start sketching your ideas.
To start, simply match some of the keywords you’ve listed above, in the answers to your contextual questions, to types of charts, summarized in the visual below.
Information source: Andrew Abela
For example, if you’ve articulated that you need to “compare” or “contrast” information, then you probably need a Comparison chart. Or if you need to display “parts of a whole,” then you’ll need a Composition chart.
One good technique to make sure you’re using the most appropriate visual format is to try two completely different approaches to see which is more effective at communicating your one main idea.
Each sketch should bring up more ideas, and once you find yourself making refinements of the same idea and thinking about colors and the actual numbers, you can start prototyping.
Below, you can find a list of free online apps you can use to do this:
Step 5: Refine Your Chart By Decluttering
Now, you’re in the final–but very crucial–stages of the information visualization process. Here is where you can use your knowledge of preattentive features to create visual hierarchy, which involves arranging elements on a page in a way that directs readers to perceive them in a certain order.
In text, we see this all the time: Big, red headlines catch our immediate attention; so do underlined, bold words or inverted elements.
When dealing with charts and graphs, color is also an effective tool for drawing attention to certain information; so is size in terms of the amount of space occupied by each element.
To ensure that your visualization is well structured and has visual hierarchy, include the following:
- Title
- Subtitle
- Visual field
- Source line
- Axes
- Labels
- Captions and legends (where necessary)
Cole Nussbaumer Knaflic recommends refining visualizations by first eliminating all unnecessary and repetitive elements. Ask yourself: “Do I need this element to get my message across?” If not, consider eliminating it. When you’ve already defined the key message you want to communicate, it’s much easier to determine what’s necessary and what’s not.
Image Credit: Visme
Then ask yourself: “Can this detail be summarized?” If your audience doesn’t need to know the specifics, then consider summarizing, where appropriate. For example, if you have 20 different characteristics than can be grouped into “super categories” without losing meaning, then simplify information this way.
Another helpful tip for focusing your audience’s attention is to push all elements to the background by applying a light gray color to them. Then go one by one, in order of importance, and deliberately highlight the most important elements by applying a different color, increasing the line thickness, increasing the size or adding labels and data markers to especially noteworthy points.
Here’s a summary of a few more important points to keep in mind when decluttering your visualizations:
- Stick to a few colors, no more than 2 or 3. And use grey for elements in the background that are not as important.
- Limit the distance the eye has to travel by placing labels and legends close to the pertinent information.
- Eliminate chart borders and gridlines.
- When possible, label columns, lines or segments directly instead of using a legend, which forces the viewer to do more work.
Examples of Popular Charts and How to Improve Them
Form Should Follow Function
Now that we’ve gone through the process for creating effective charts and graphs, let’s put this to the test by critiquing a few of the most shared information graphics in the past year.
Take a look, for example, at this extremely popular infographic on the most deadly diseases in the U.S. compared to those that attract the most donations.
Image source: Vox
Besides the fact that the size of the bubbles are not accurately proportioned in relation to each other, it’s also important to note that this is not the most effective visual format for displaying this type of data. Since the human brain is not as adept at calculating differences in area as it is in comparing single dimensions like length or width, using bubbles to make comparisons can be misleading–even if they are beautiful.
The design team at Visme recreated this into a horizontal bar chart and came up with this:
Side-by-side bar charts are a much more accurate way to present this information because they all start at the same baseline. Notice how the magnitude of the difference between values is easier to discern when comparing only the length of the bars. Here there is no need for a legend, whereas in the first version viewers were forced to look back and forth between the bubbles and the corresponding labels.
Use Slopegraphs to Compare Rate of Change
Another example is this widely shared bar chart displaying the number of young people who voted in the primaries of the U.S. elections in 2008 and 2016.
Via Washington Post
Although this chart is not terrible, it could be better. One way to improve it is to use a slopegraph instead of a bar chart. While the latter is ideal for displaying categorical data, the former is useful for displaying the rate of change between two points in time, visualized through the slope of the line.
Push Secondary Information to the Back
This next graph is a perfect example of how too many colors can distract from the main message. The point is that since mid-2014, the supply of oil has exceeded demand, resulting in excess (blue bars).
Via Vox
To fix this, we pushed everything to the background by applying grey to all the elements and then started to bring to the forefront the most relevant elements, such as the change in supply and demand starting in mid-2014.
In this second version of the chart, we see that the axis values, title and subtitle are deemphasized, while the relevant information is highlighted through the use of color.
Let Us Know What You Think!
Do you have any chart makeovers or data storytelling tips of your own you’d like to share? Leave us a comment below and let us know your thoughts.
About The Author:
Nayomi Chibana is a journalist and writer for Visme’s Visual Learning Center. She has an M.A. in Journalism and Media from the University of Hamburg in Germany and was an editor of a leading Latin American political investigative magazine for several years. She is passionate about data journalism and researching trends in interactive longform media.
Twitter handle: @nchibana
*Featured Image Source
