Innovation in Data Visualization

There's been a lot of recent posts and debates within the data visualization community lately and one of the recurring themes is innovation, or as some have claimed a lack of innovation in the field. I'm not going to enter that debate at the moment, but instead talk about the process of innovation. People use this word quite frequently. Listen around you and I think you'll be surprised at how often people talk about it. But how do you actually do it? How do you innovate?

The first thing that probably comes to mind is "brainstorming". Put a bunch of people in a room and start ideating and then magically these innovations will come popping out, the proverbial light bulb moment. Unfortunately, for anyone that's done this, and I'm sure that's most of you, it's often a painful process that frequently fails and doesn't lead to true innovation. Think about it. If I said to you, go off in that room with these 3 people and come back with a new method to visualize data, for example a new chart type, what would you do? How would you approach it?

I am honored to be able to teach Data Visualization at the University of Cincinnati in the Lindner College of Business. We have some brilliant professors across many disciplines. One of those professors is Drew Boyd, a true thought leader on the subject of innovation. I've had the opportunity to hear him present on a number of occasions and I became an instant fan of his work. Why? Because Drew teaches a "system for innovation". Instead of approaching innovation as a random thought process, he teaches Systematic Inventive Thinking (SIT).

What is Systematic Inventive Thinking (SIT)?

Systematic Inventive Thinking is an approach to creativity, innovation and problem solving. It's a methodology that can be executed as a process. Instead of relying on random ideas, specific steps can be followed through a defined process. What data person wouldn't love that?

Drew describes the history of the SIT method coming from the research of his friend and co-author, Dr. Jacob Goldberg. This research utlimately led to the thinking tools that are applied today in SIT.

I'm not going to discuss the entire SIT process in detail, as I'm certainly not an expert in this area. For those of you who would like to learn more, I would encourage you to read Inside the Box: A Proven System of Creativity for Breakthrough Results by Drew Boyd and Jacob Goldenberg available here on Amazon. Drew also teaches a course on innovation on Lynda.com and be sure to follow Drew on Twitter here.

I do want to share two of the fundamental principles of SIT and the Five Thinking Tools.

A Few Principles of SIT

The Closed World Space - Instead of the old adage "think outside the box", we actually need to "think inside the box". We don't want to have to rely on some new way of thinking, something completely different from what we've done in the past. We approach innovation with the problem at hand. We can only solve that problem with the tools and components that are in front of us. This causes us to focus even more on the details of the problem and the components.

Function follows form - If you have studied data visualization, then it's probably been grilled into your head, "form follows function". And for creating data visualizations, that's still a great rule to follow. However, when innovating, we need to flip this around. It may seem counter intuitive at first, but I'll explain this more in a moment.

The Five Thinking Tools

1.) Subtraction - remove an essential component to the process and then examine what remains. Find uses for the newly created form of the product.
2.) Multiplication - adds something that exists in the current form, but change it in some manner.
3.) Division - divide the product or components into pieces and rearrange them
4.) Task Unification - assign a new task to an existing component of the product
5.) Attribute Dependency - creating and resolving dependencies of the variables

This is a lot of information to digest, so rather than diving deeper into each of these, I will apply this process to data visualization to demonstrate how this might work to generate a new way to visualize information.

Applying SIT to Innovation in Data Visualization

I will start with the standard line chart. This chart is the go-to chart for plotting time series data. It typically has an x-axis for time and a y-axis with a measure. For this example, I will use the Superstore data from Tableau and plot sales over time.

Now let's apply the SIT method. I am going to use subtraction. Remember it's a closed world space. Everything we use is already given to us, so we aren't going to add any new components to our chart. Instead, we are going to subtract something from the chart, something that is a very important component. Let's drop the x-axis and the y-axis and remove the title. A chart's axis is a very important component, after all, it gives the user an indication of the time scale and the dollars of the sales. Notice that when we go through this process, we are not trying to solve a specific problem. We're making the changes to the form before we consider its function. Here's what we're left with.

Now that we've made these changes we consider function. We ask ourselves, what can this be used for? how is it useful? what purpose can it serve? And after doing that here, you might come to the conclusion that, "if we reduce its size, then it shows the trend of sales over time in a really small space".

Hey, I just invented a sparkline! Well, Edward Tufte beat us to it, so we can't take credit for this one. Indeed, this sparkline can be useful. It doesn't tell us exact numbers or dates, but when reduced, it becomes a dashboard component that is information rich in a word-sized space. Let's apply one more, multiplication. We add more lines, but make a change to each one. I'll change one line to profit instead of sales and another to quantity. And then we evaluate it again. Examining our three lines, we not only see trends over time, but we can see the interactions between them. Again, in a very small space, which can be very useful in dashboards or on mobile devices where space is very limited.

Once we've reach this point, we can think more about the function it serves, and make improvements in the result to create the new finished product. For example, adding an average line and an end point with a label. We've now improved our new invention to create something that is even more useful. But notice, we didn't add anything during the innovation process, only after we had a new product.

I would encourage you to take some time and learn more about Systematic Inventive Thinking and how it might apply to the field of Data Visualization. I believe that this powerful framework will enable us to innovate new and useful visualization techniques, ultimately helping us to advance the field of data visualization.

You might be wondering at this point why I'm not off innovating more with this approach. Unfortunately, I have not had the time. I have been working, teaching, writing a book and trying to keep up on blog posts, so hopefully the community will find this information useful and give it a go.

I hope you find this information helpful. If you have any questions feel free to email me at Jeff@DataPlusScience.com

Jeffrey A. Shaffer
Follow on Twitter @HighVizAbility