“Organizations continue to push data-driven decisions for greater amounts of their activities. As a result, the demand to understand meaning within the mountains of collected data is rapidly growing. The bottleneck in understanding all this data is the scarcity of individuals (read: data scientists, statisticians, etc…) who can use this data to glean meaningful insight. This has historically been a challenge for many organizations, but data visualization tools can help us bridge this gap.” James Haight, Blue Hill Research
I love this quote. As one of those statistician/data scientists the author speaks of, I can testify to the growing need and demand for effective data visualization. This idea is not lost on us leaders either, as more and more we have to influence people through data. More and more data gets captured in more and more systems, from simple spreadsheets to complicated software platforms. According to Andrew McAfee and Erik Brynjolfsson in their October 2012 article Big Data: The Management Revolution, as of 2012 about 2.5 trillion gigabytes of data are created each day, and that number is doubling every 40 months or so. More data crosses the internet every second than were stored in the entire internet just 20 years ago. To put that in perspective, think about this comparison. Let’s say that water bottles were the equivalent to a byte. That would be 1 billion bottles to a “giga-bottle”. So, 2.5 trillion “giga-bottles” would be 2.5 quadrillion bottles. If each bottle were eight inches long, you could put that many bottles end to end and line them up from Earth to Pluto and back 32 times…and Pluto is 4.67 billion miles away! That’s a lot of data!
Why does this matter? As McAfee and Brynjolfsson point out, it’s because data driven decisions tend to be better decisions. The question we should always ask ourselves is not “What do we think”, but “What do we know?” The effective capture, analysis and distribution of data and information allow us to do that, and to take the guesswork out of decision making to the best degree possible, whether we are data scientists or leaders. Companies spend a lot of time and resources on systems that capture data, and it seems like everyone wants a piece of the output…but they want it in a way that they can easily understand, and that often means a graphical display of some sort. On the surface it’s easy enough to understand why. The human brain works well at picking up patterns, so any sort of chart or graph that helps highlight a pattern is something we can easily understand, and quickly interpret with minimal instruction. From a pie chart to a scatterplot, we’re good at seeing what we need to see when it’s in graphic form. For leaders who must influence others through data, this is a must these days.
The challenge for those of us who work with data intended for presentation is to create visuals that achieve this. We might have extremely complex data sets we work from, but still need to create output that tells the story we need to tell in a way that’s easy to grasp, clearly understood, and visually appealing.
Blue Hill developed “The Cognition Matrix”™ that neatly illustrates this challenge. Basically, it describes the process of “pushing” data from a complex form to a simple form, by taking advantage of the way our brains process visual data in a natural, intuitive way.
On the vertical axis of this matrix is the complexity of the data being presented. Calculation is simple; for example, the average scores for program satisfaction for a CCL program designed for senior leaders is a simple calculation; we learn to calculate averages in grade school. Computation at Scale, however, is complex; doing a comparison of means across multiple programs to look for significant differences in satisfaction, for example, isn’t something we can just “see” by looking the data. We have to compute it to get to it.
On the horizontal axis is how we process data; on the left is Binary Processing, or “thinking”. On the right is Pattern Recognition; think of that as “instinct”.
So the goal of any data analysis should be to push information from the left side of the matrix to the right, where instinct takes over and helps to interpret the data. Even very complicated analyses run in the upper left quadrant can be filtered down and displayed in a way that would make the output fit in the lower right.
How does that work? Consider this example from Blue Hill about catching a ball. If we are thrown a ball we could figure out the calculations behind how to intercept and catch it, but we don’t. We use instinct and past experience to know where to move to so that we can catch the ball. The same goes with understanding data; if you show data in a familiar pattern, the audience will be able to understand it without much need for help. They might need a bit of context to understand the importance of the information, but understanding it is something they can do by simply looking at it. This can be difficult to do for data analysts and leaders too.
So what does this mean for you? Well, I suggest adding one more facet of detail to this model to help you think about how to display information; Presenter and Audience. The presenter is YOU; the person working with the data and preparing it for dissemination. You spend your time on the left side of the model, where all the processing and cognition really takes place. That’s your area, because it’s your data; you understand it on a more detailed level because you own it or work with it. But on the right side is the audience, or the people you plan to share the data with. They are likely not as familiar with it, and in many cases have limited time to look at what you’ve found.
The challenge of any data presentation a presenter might prepare should be to use this model, and try to move the data from the complex side of the grid to the simple side for your audience. You might have to spend some time manipulating your data, consolidating it and condensing it to get the type of output you want, but in the end, that’s your job – to show the data to your audience in a way that tells the story you need to be told in a clear, effective and easy to understand way. The purpose of this blog series on data visualization is to help you do just that.
With this blog as the kick-off to this series, we’ll talk about how to create effective visuals, balancing the density of the information with the visual appeal and ease of understanding so that you can make the best visuals possible for your audience. We will even delve a bit into how the brain processes data, so you’ll know a bit more about how people will react to your graphics, even when they don’t realize it themselves. And, we’ll do some step-by-step instructions on some graphic generation to help you build the types of graphics you want to show. In the end, no matter if you are data scientist, a data junkie, a leader who wants to display your data better and convince others of your data-driven decisions, or any combination of the above, we hope you’ll be able to take the lessons provided and create more effective data visualizations for your audience.
Haight, J. (2014, October 13). How Data Visualization Empowers Decision Making and Who Is Getting Us There (blog post). Blue Hill Research. Retrieved from http://bluehillresearch.com/how-data-visualization-empowers-decision-making-and-who-is-getting-us-there/
McAfee, A., & Brynjolfsson, E. (2012). Big Data: The Management Revolution. Harvard Business Review. Retrieved from https://hbr.org/2012/10/big-data-the-management-revolution/ar