Diving Deeper

Making Data Dynamic

Using Data Visualization to Tell Your Story

We all use words to communicate. And often, in an effort to enhance the impact of our message, we rely on nonverbal gestures – hand waving and eyebrow raising, smiles and frowns. The dynamic relationship between word and gesture not only facilitates communication, but it also improves audience cognition. Research shows that gesture not only contributes “essential information to a message but also actively facilitates the cognitive formation of messages and . . . communicates unique information that is not present in the speech signal.”

The same dynamic relationship exists between data and data visualization. As David Buckler discussed in his three-part blog series, “Making Big Data Smart Data,” we’re awash in information. A 2017 study by IBM revealed that 90 percent of all the data in the world had been created in the preceding two years. Most of it is siloed into databases and file systems, holding the secrets that could translate into more profitable products, better customer service, and more efficient operations.

By creating graphical representations of your data, you are translating information from a computer readable format into the human realm. You are using numbers and data to create a story that provides your audience with access to fresh insights. We know a picture is worth a thousand words. That maxim holds true for data, too.

Data visualization is the representation of data in a visual, or graphical, format. Done well, data visualization tells a story, providing key insights that help users understand trends, recognize patterns, identify outliers, and, ultimately, make business decisions. Done poorly, it overwhelms with information, misdirects attention, and leaves users hunting for answers when trying to make business decisions.

Start with Audience and Purpose

Because so much information is continually being generated and people know this data holds valuable insights, many organizations are inclined to create an abundance of charts, tables, images, and graphs as quickly as possible. In doing so, they’re creating unreadable, often cluttered dashboards that provide little in the way of a clear message.

The reason you’re creating a visual is to tell a story, and to tell the story, one that has practical value, you have to know what your audience needs to know or wants to learn. It’s helpful to start with the right questions when you discuss the story with your audience:

  • What is your end goal?
  • What answers are you seeking?
  • What decisions do you need to make?
  • What specific outcomes are you trying to achieve?
  • How do those outcomes translate to the business problem you’re solving?

There’s some trend, pattern, or insight you’re trying to interpret or uncover that influences what you’re collecting and how you present it. You’re not measuring or tracking data simply to measure. Rather, you’re measuring to influence someone who needs to make a business decision.

Understanding the intent and use case before you start creating dashboards and understanding how the information will be used, you avoid the risk of overloading people with data they might not need at that moment.

Tailor Design to Audience and Purpose

Just as you want to create graphic representations of data based on the story your audience wants to convey, you need to tailor your use of graphic elements so that your visualizations are engaging and compelling. In other words, aesthetic form needs to align with practical function. The key to good design is to be simple and intuitive. It’s important to initiate the process of design with an answer to the question, “What are you going to do with this dashboard once I give it to you?”

Here are facets to consider when designing data visualization instruments:

Types of Data Visualization

When it comes to conveying your message graphically, it’s important to pick the right graphic for the job. There are lots of choices, including charts, tables, line graphs, infographics, timelines, maps, and pie and bar charts. Each type of visual serves a distinct purpose, and it’s important to choose the right one for the specific job. Line graphs, for instance, reflect change across time from one data point to the next. Customer growth is one example. Bar charts are effective when you’re comparing categories within a certain measure. In this case, you may be showing the profitability of one credit card over another. The goal when selecting the type of visualization is to make sure your audience gets the message it needs with a quick glance and that you are communicating the most important information.

Color Selection 

While some people may think a rainbow effect makes a graphic attractive, color should be used sparingly. By focusing your use of color on the story you want to tell, you are more successful in your effort to highlight the data and information you want to communicate. Color is sometimes the most important design element to think about because that’s the piece of the message that draws in your audience’s eye. Colors signify different things for different people, for instance, based on age, geographic region, and culture.

Here are a couple points to note. The colors you select initially to convey various points in your message become set in the mind of your audience, and they may interpret that color in the same manner with any visualization you add to the interface. So be consistent. Also, you need to consider whether your visual is showing your audience something you want them to see or something they need to see. Each context might require a different color.

A Living, Agile Product

As diligent as you are in determining precisely what your audience wants and needs to know, you often get a clearer understanding of their needs only after they begin to use the data visualization tool you create. Interacting with the data visualization tool you develop, users will tell you as they click through layers of information what they would like to see in the next iteration. You don’t want to start off by giving people information that’s too granular because it won’t interest them, or it will distract them. The key to maintaining usefulness and relevance is to let people work their way through your visualizations and then use their preferences to sharpen the focus of the story. When you do, you improve the effectiveness of your dashboard with each iteration.

When to Step It Up

 Much of the time, a dashboard tells a tried-and-true operational story, where standard bar or line graphs and some drill-downs are plenty. When consistency or familiarity is key, over-engineering can get in the way. For a complex, open-ended question, however, you may need more dynamic reporting and visualizations. If your board relies more on standard summary views, stick to that theme. When users need to dig into something like a multivariate problem, explore geography, or understand hierarchical mapping, it’s time to get fancy. In the most complex cases, a simplified dashboard with interpretable machine learning might tell a story for those who need it while keeping things consumable for the average user. Know your audience.

At the end of the day, it’s important to remember that data visualization should summarize and then present information in a form that is engaging and easy to understand. Cognitive psychologists use the term “Goldilocks effect” to explain how infants are inclined to attend to events that are neither too simple nor too complex. That concept might just apply to data visualization. Your goal is to simplify, but not oversimplify, the issue at hand. You do that by answering questions for your audience to help them make decisions, and you present your information in the appropriate graphic using intentionally selected design elements. And you don’t want to create complexity with superfluous information and overwrought design. When you are “just right” in the way you democratize data, people will have the autonomy to use information easily enough to be an expert in a particular set. That’s good business. It’s also empowering.