The Rise of Data Visualization
Hello avid data vizzers! Steven here, bringing you some lightning-quick tidbits to fill your brain with interesting knowledge that’s also good conversation fodder for parties. (Depending on the parties.)
First, a very very brief history of data visualization, summarized from the already brief “A Brief History of Data Visualization”:
Way back in “the day” (at least as far back as 200 BC), we have evidence of people working on large physical problems like plotting planetary movement, preparing city layouts, and creating maps of the known lands. and routes for ship navigation.
With large physical problems came the need for the measurement of physical things like time, distance, area, etc. Cartography continued being an area of great interest, especially as nautical exploration and trade became more prominent.
William Playfair invented many of the charts and graphs we use today, including the line graph, bar chart, pie chart, and circle graph.
The study of demographics and qualities of groups of people led to social science, crime reporting, disease tracking (like John Snow’s famous data viz of cholera), and economic and state planning.
A few concentrated periods of statistical advancements greatly contributed to the math behind visualizing data.
The “Dark Ages of Data Viz”, first half of the 20th century – this was a time when the precision of quantitative models was valued over “imprecise” visualizations, and few new data viz developments were made.
The second half of the 20th century brought the revitalization of visualization with the introduction of computers and the continued development of applications in other scientific fields.
An early “small multiples” visualization of sunspots.
Google Trends for Data Visualization
This brings us to the modern day. First, let’s take a look at Google Trends data for the last decade.
(Note: the vertical axis here is relative to what the highest point is; since the top right point had the greatest traffic, that is set to 100% and everything else is plotted accordingly.)
“Data Science” has gotten a decent uptick in the past few years, probably because of the massive amount of press it’s gotten as being the sexiest job around.
If we throw Tableau into the mix (yellow – the current leading data visualization software), it massively outperforms both of the other search terms.
(Also, every single one of those big valleys that you see is “July” of the respective year. Why July? It’s the beginning of Q3 in the business world, and perhaps that’s always the low point in the sales cycle for whatever reasons.)
Tableau (blue in this graph) is also far above any of the other popular visualization tools out there right now.
Let’s look at the frequency of the phrase “data visualization” throughout all of the books Google has digitized and made searchable via the Ngram Viewer.
We can see an exponential growth of references in the last 50 years or so, but there’s one other blip that’s incredibly interesting…What happened around the year 1800?
(I’ll leave that question unanswered so that you can have fun doing some Google hunting.)
After hundreds of years of growth and developments across the world across a huge range of applications, we can see that data visualization has really exploded in recent years thanks to the help of computers.
Why is data visualization important? Well, let’s talk about human evolution. Our brains have evolved over tens of thousands and millions of years to make sense of visual sensory input (and to do so very quickly, as our lives may depend on it), but before the past few thousand years we didn’t have too many natural reasons to develop a numerical computing region of our brains – basic counting was probably all we needed. Thus, we’ve evolved to be very good at visualizing things, but pretty bad at numbers. Taking in and processing visual input is a computationally intensive task, too, so a huge portion of our brain is used for or affected by vision.
Parts of the brain used for or affected by vision.
To process numbers, our brain first has to convert the image of the numbers into the idea of the numbers, and then we can start our sloppy inefficient mental computing. If you don’t think you’re bad at numbers, get a spreadsheet of at least 100 different numbers and try adding the whole sheet together in your head. The majority of people will make at least one small error, and even if they come up with the right answer it will take minutes of intense concentration spent adding the numbers one by one. (Unless you’re someone like Daniel Tammet, who is an incredible anomaly.)
What is most of our data in this modern society we have constructed? For a lot of businesses and organizations, data is numbers. Lots of numbers, organized in rows and columns in massive databases or spreadsheets. Research labs often generate images, video, sounds, or other types of information like that, but even these can often be represented by numbers. Unfortunately, like I just pointed out, we’re particularly bad at numbers.
What are we good at? Vision.
So what should we do? Visualize our data.
Data visualization? Bingo.
Science break over.
The Modern Power of Data Visualization
Tableau made it their mission to bring data visualization to the “anybodys” of the world, and they hit something big with that approach. Of course, as soon as others saw how well Tableau was doing, they wanted to hop on the data viz bandwagon as well. Lukas Biewald has a large breakdown of the current data science ecosystem that we see today, and many of the tools listed have started implementing their own visualization solutions primarily in response to Tableau’s market domination. At Boulder Insight, we’ve seen time and again the power of making interactive visualizations accessible to everyone in an organization.
(Side note: around the same time period that Tableau hit the data viz scene, infographics started exploding on the internet and are now mainstays through sites like the KISSmetrics blog. Although static, infographics allow much more visual creativity and are much more engaging than standard text-based presentation of information.)
Infographic results of the 2010 British Elections.
There’s a reason that data visualization has been around for so long, and it ties deeply into who we are as humans. From cave paintings, to the ceiling of the Sistine Chapel, to comic books and sunsets, we rely on and are enamored by the sights we see. Our eyes bring us inspiration and information. Even ridiculously simple images like those found in XKCD comics are enough to engage people.
(title text: “You can do this one in every 30 times and still have 97% positive feedback.”)
Knowledge is only powerful if you use it, so here’s the main take-away for you to try: challenge yourself to use beautiful and informative visuals whenever and wherever you can. Use them in presentations, use them with your kids or parents, use them to tell stories, use them to brainstorm. Our brains have been wired for vision, we enjoy it, and you’ll find incredible success when you utilize it well.
Happy visualizing, friends.