Watching my daughter occasionally play Birds and Blocks or other games on the iPad causes me to ponder the level of interactivity and visual design we have come to expect in so many interfaces — the internet, mobile devices, PC apps, dashboards and more. In an earlier blog post on graphicacy levels rising, I wrote that we have become more visually literate and explained some of what’s contributed to that. I contend that this trend will continue in 2013, and furthering my argument, I also predict greater adoption of interactive and dynamic visualization. This requires a closer look at what many people are using today to graphically encode data (primarily static graphs from spreadsheets) and why I think they will gravitate to using more interactive and dynamic capabilities—both to understand their data and to communicate results.
Getting to know the data is key to gaining insight and realizing value. More dynamic and interactive graphics greatly facilitate the ability to more quickly explore and understand the data’s structure, patterns, relationships, trends, and anomalies, which is even more important in this era of big data. By marrying the power of high-speed computation and displays to render such dynamic and interactive graphics with our high-speed visual bandwidth to navigate them, we gain several benefits.
First, analysts are empowered to carry out the exploratory phase of analysis more flexibly and can more quickly get to know the data, speeding time to insight and value. One could also argue that having such flexible and fast, visual discovery tools, encourages analysts to more deeply explore the data. The dynamic and interactive aspects of good visual exploratory data analysis make it so easy (and even fun!) to keep asking questions of potential interest that you increase the opportunities to find more interesting things you weren’t necessarily seeking. John Tukey, the father of Exploratory Data Analysis said, “The picture finding eye is the best finder we have of the wholly unanticipated.”
Second, analysts can stay “in flow.” Dynamic and interactive graphics allow analysts to — at the speed of sight — keep unfolding the analysis in a way that suits their analysis styles. Many may not realize it, but analysis is personal — I may want to explore outliers first after seeing a time series plot, whereas someone else may want to explore seasonal patterns. What one finds most interesting to explore further will vary from person to person as will ideas about what to try next in the analysis. Through visual paradigms — dynamic filtering, linking, column-switching and various other means — the analyst can stay in flow, trying things on for size with minimal effort, to see relationships, patterns, anomalies and more to inform next steps in abstracting knowledge from data.
Staying in flow allows analysts to do what they enjoy most: analysis. It minimizes the drudgery of having to write code for tasks that are best automated or having to go back to a template and change a setting to rerun the analysis. Data should work for you. You shouldn’t have to work so hard to make sense of it when the capabilities to visually navigate the data and more easily reveal insights are there. That said, there is always a place for code, but even hard-core coders can be more efficient with visual paradigms added to their analytic workbench.
Third, having become more intimately familiar with the data, analysts can more easily convey findings to others through interactive, dynamic presentation (versus discovery) graphics. The data’s stories can often be more compellingly conveyed dynamically, interactively — even playing “data movies” — making the results more memorable and compelling. True story: several years ago I presented to a mixed audience at a large bank showing a data movie of crime trends in the U.S. A year later, I was meeting with an executive who was interested in seeing data movies of their own data—he began to describe what he had seen a year earlier. Clearly, he remembered the story, though not the presenter. It comes down to visual storytelling, making the narrative in the numbers apparent. From Steven Denning’s The Leader’s Guide to Storytelling, “Storytelling doesn’t replace analytical thinking, but it can supplement it. Analysis might excite the mind, but it hardly offers a route to the heart.” Most analysts want their work to have more impact. Distilling the data into compelling stories that people can see can have greater influence.
Finally, information consumers can more quickly see and understand results of the analysis. Two important things flow from this. First, new questions arise, questions that may never have come up if not for seeing and understanding the story: “Oh, why is…? What if…?” Second, the visualization sometimes reveals truths that help overcome previously held unfounded beliefs, expectations or biases. Putting the data in the right visual context helps others “see” the relationships, interactions, etc. Without that visual context, the results may be summarily dismissed. Accepting new insights and asking more (good) questions leads to better decisions. Interactive and dynamic visualization speeds this process.
These benefits are becoming more apparent, driving greater adoption of more powerful visualization capabilities. In addition to the previous blog post on graphicacy, I’ve seen more evidence that adoption will grow: Alberto Cairo, author of The Functional Art (very enjoyable book), through the University of Texas at Austin, is offering a free online course on infographics and data visualization for a second time because the first one that debuted a few months ago was so popular.
More books on interactive and dynamic graphics have been published in the past few months, and more are due out in early 2013. More universities are incorporating the power of visualization in course work and exposing students and faculty to events about visualization. Even K-12 schools are looking at using iPads with free apps like the JMP Graph Builder app to incorporate more hands-on, interactive visual learning. Because the learning curve is so fast and visually intuitive, with better designs and usability, I’m optimistic that more people will soon understand more about their data through dynamic and interactive visualization.