Balancing Analytic and Visual Creativity for Effective Data Visualizations
“The greatest value of a picture is when it forces us to notice what we never expected to see.” – John W. Tukey
It has long been said that beauty is in the eye of the beholder. That’s an adage that rings especially true in the quest to derive data-driven clarity from raw or analytically-synthesized results and where achieving insights through visualization requires a unique combination of quantitative discipline and judicious creativity.
Data visualizations are highly effective tools for communicating the power of data and analytics, as they showcase results in easily digestible forms that can be tailored to audiences with different learning styles and capabilities. Optimizing the communicative potential of visualizations requires drawing upon the unique skills derived from both the “right and left brain”. It has been said that the right brain is one’s creative and intuitive side, while the left brain represents the logical and analytical side of one’s persona. Regardless of the scientific legitimacy of the left- and right-brained dominance theories, the concept reinforces the fact that those who create data visualizations need to be able to think in two very different ways to communicate effectively to diverse audiences.
Know your audience
Visualizations tailored to their audience and created using carefully selected visual and analytic techniques can improve data-driven decision making by end-users. To realize increased business value, the visualization needs to be communicated at the right time, in the right format, and in the right amount. Different circumstances and contexts call for different visual approaches.
Visualizations tailored to their audience and created using carefully selected visual and analytic techniques can improve data-driven decision making by end-users.
As well, the traditional interrogatives should be considered: Who is the audience? What do they want to know or what does the visualization analyst want to convey? When are the different formats most appropriate? Where, and in what order, should information be structured for maximum impact? Why does the reader need to know or understand the certain content? How does the reader want/need/prefer to be presented information in a way that best meets their learning style?
Every audience, knowingly or unknowingly, has different preferences for viewing and comprehending visualizations, whether simple or complex. The visualization analyst is challenged to create a unique presentation that aligns to the audience’s learning style, allows them to absorb information quickly, and increases overall understanding. Some people are visual learners and prefer the “thirty-thousand-foot view.” Others prefer a moderate level of detail, while some want to dive deep into visuals and quantitative detail. Audiences may be top-down learners, tending to learn best when they start with a summary, then move to the fine details, or the opposite, bottom-up learners. These characteristics are important to consider when designing visualizations for maximum impact and efficacy.
Use your right (creative) brain in visualization development to:
Tell a story. By harnessing the power of the right brain and using insights gleaned from knowing the audience, the visualization analyst tells a story using data. For example, pairing different visualizations together in a dashboard fashion may foster a sequential discovery of the facts. Exploratory visualizations through techniques like drill-downs can tell a different story at a different pace than static explanatory visualizations. Multi-dimensional visualizations (visualizations user can interact with by adjusting parameters) can tell a different story than canned visualizations. Balancing flexibility with complexity to avoid overwhelming the user is essential; too many choices can cause the user to lose interest or feel overwhelmed. Balancing the concept of “just the facts” with “all the choices” are important decisions to be made.
Pick the type of visualizations. Any visualization opportunity offers dozens of types to choose from— heat maps, tree maps, bubble charts, word clouds, stacked graphs, spider charts, and more. Dr. Andrew Abela is well known for a published primer on how to select a chart type with the logic centered around four display criteria–comparison, distribution, composition, and relationship.
Most data visualization types can be categorized in this fashion, and having an understanding of this decision logic is helpful in selecting the desired visual. Once you decide on the data story you want to convey, creativity is necessary to choose the visualizations that can tell the story most effectively. Factors may include the type of data you have, the time period of the data, your messaging, and your “know your audience” parameters.
Create a visually appealing and effective design. Right-brain thinking is also required to manage the specific design considerations. For example, the analyst must consider how visual specifics— position, sizes, colors, shapes, patterns, and density—may affect the interpretation of the results. A visualization analyst should avoid excessive embellishments and too many different styles and patterns to help capture and allocate the scarcest resource a decision maker has – their time and attention. Be aware of the potential for audience absorption exhaustion: the dread a user feels when each turn of the page is a new visual experience that requires an ongoing investment of time and mental energy that can result in a shutdown and dismissal of the material.
Use your left (analytical) brain in visualization development to:
Understand the data and maintain quality. Left-brained thinking is required to ensure data quality in the visualization. The analyst needs to think programmatically to clean, prepare, and normalize the context of the data and ensure its completeness, validity, integrity, consistency, timeliness, and accuracy. Left-brained systemic thinking helps identify potential risks in the data and methods to avoid gotchas. A detailed understanding of the data structure is also necessary. How does one group of the data link to another? Is the data categorical, ordinal, relational, time-series, one-to-one or many-to-many, linear or non-linear, etc.? These and many other detailed data issues must be explored and understood.
Maintain visual consistency. Creating visually consistent displays is important and requires logical planning and design. Both companies and analysts need to recognize that if they are going to be using visualizations frequently, designing some standards of use, templates, and stock-types, is a strategic decision for consistency and user acceptance.
Finding a whole brain thinker is challenging. Visualizations are the future of presenting, not simply a new or better way to present. Visualization analysts who have an analytical skillset—coupled with artistic literacy—may be difficult to find, but they are essential for creating effective visualizations. Team members who have the capability to harness both lefts- and right-brained thinking and the necessary technical skills should be identified, developed, and nurtured to become your organization’s analytics storytellers of the future.