Data Visualization: The New Literacy
Say the words “information visualization” or “data visualization” these days, and you’re likely to get someone’s attention. In today’s world of big data and nearly constant information overload, these kinds of visualizations are everywhere and are often seen as a sure shortcut to insights. Are they? Sometimes, but not always. While it’s true that information visualization is a great tool, it is not a silver bullet. Visualizations facilitate understanding, but they should be considered within the larger context of storytelling and effective communication. They can be good and they can be bad, just like anything else.
Designing and understanding data visualizations is a skill, and graphicacy is rapidly becoming a new kind of literacy. You don’t have to be a graphic designer to create effective data visualizations, but you do need to know some basics to communicate effectively through visualizations. So what are those basics?
First of all, we need to understand what information visualization is best used for. Its benefits ultimately bubble up to three key values: seeing the big picture, recognizing patterns, and making comparisons. Why those? When we need to see the big picture, visualization wins over text for its ability to give us a quick, holistic understanding of the entire system we are investigating or explaining in a way that simple text could never show us. For pattern recognition, visualization taps into the phenomenal capability the human brain has to respond to visual inputs and pick up on similarities and differences. The same capability aids comparison: while verbal descriptions can achieve the same accuracy of information, visualizations greatly help understand the distance between the values as well as the overall scale on which the items are being compared.
Secondly, it’s important to understand how we can do it better, even with the existing skills we have. Here is where people-centered design comes in. Think of information visualization the same way you think about a product. If you’re someone who practices people-centered design, you would start designing your product by understanding and empathizing with your audience, then you would brainstorm concepts that respond to their needs, then test and iterate, and so on. Let’s look at how information visualization can benefit from that first step: understanding your target audience.
What are the key things you need to know about your audience? When designing data visualizations, we recommend you consider people (who they are), context (what the environment is like where they operate), and objectives (what they are trying to achieve).
Who are the people who will be looking at your visualizations? What is their background? What is their knowledge of different visualization conventions?
For example, every scientist knows how to read a box plot, since box plots are a very basic, common type of scientific visualization. Ask a non-scientist with the same level of education, and you are likely to get a blank stare in response. Some data visualization gurus like Edward Tufte believe that if a visualization is a good one, anyone will be able to understand it, but our experience challenges that view. Considering factors like your audience’s prior knowledge, experience, and expectations are likely to lead to more successful communication.
How will the visualization be presented and consumed? How much time does your audience have to process it? How much context do you show?
It’s one thing to have the time to invest in diving into the details of a complex map, it’s another when that same map is flashed on a screen for a few seconds. When you consider context, keep in mind the medium through which your visualization will reach the audience, the size of the visualization, and the time it will take to process it. If you present a dense visualization in a meeting on a screen, consider building up your visualization over several steps. For example, introduce main elements first, then add connections between them, then add details to the elements. This will help ‘unfold the visualization’ and ensure that your audience gets it. If you leave the same dense map behind, consider making it printer-friendly as a large poster that can be put on the wall.
What kind of story does your audience care about? What are they trying to achieve with these graphics? Think about the three most common objectives of visualizing information: to explore, to explain, and to decide.
Data exploration is about finding out what the story and its elements are. Here, information visualization helps discover insights, patterns, and significant facts to help form an opinion about the subject. Some infographics fall into this category, as do many plots and tables. Interactive – as opposed to static – information visualizations and tools greatly facilitate exploration, so it is important to understand the types of questions your audience may want to ask to support that interaction.
This is probably the objective we think of most frequently when it comes to information visualization. Explanation is about understanding not just what happened but why and how it happened. Here, we use visualization to help find details and proof points and then explain them to others. For example, these visualizations may show historical context, data trends, or be embedded in other media, including text, to tell a compelling story.
This objective is about using data visualizations to support decisions and commit to a certain course of actions. Scorecards and dashboards are the most common examples of decision-making visualizations. They allow their users to quickly triage rich data to know what requires attention.
Great visual communication weaves together graphics, tables, photos and text that work together to tell a story. Even small investments in getting better at it will pay off greatly. If we’ve convinced you that understanding your audience is important, make sure to follow the other steps of people-centered design to improve your visualization skills. To support your learning, here are some of my favorite references. For those who tried classic information visualization textbooks but found them overwhelming, we can recommend simpler but equally good places to start. Know thyself to decide what works best for you, and good luck!