Why a visualization is worth a thousand data points
Jul 8, 2019
We talk about data visualization (or “viz”) a lot at Looker. We believe in its power to tell stories, to help people see patterns and take action accordingly. In other words, visualization matters—a lot.
In the age of big data, visualization is one of the most powerful tools in your arsenal as an analyst or data enthusiast. It can so drastically affect the way we interpret and understand information that it may seem like magic; when, in fact, it works so well because of the way the brain processes large amounts of information. Get it right, and you supercharge your efficiency, your ability to draw connections, and your communication.
Data visualization makes you more efficient
Understanding the scope of data available today illuminates why visualization makes you more efficient, and how necessary it is for modern business.
For example, this picture, taken on my cell phone, occupies 2 MB of storage space.
Image taken in the Sierra Nevada
Now, suppose that 2 MB could be represented by an inch (~2.5 cm) of physical space. If we laid out all the data in the world (33 zettabytes, according to the International Data Corporation) in a line, it could wrap around our planet about 9.5 million times.
It’s unlikely you’re dealing with all the data in the world at your day job, but chances are the amount of data your company collects is still impressive.
At Looker, we have roughly 15.5 TB of data (or 4.8 marathons’ worth, keeping with our earlier analogy). One particular table contained 20,687,442,124 rows of data at the time of writing this post. Making sense of a table like that it is a pretty tall order.
What I can make sense of is seeing those data points plotted on a chart, where trends are rendered visible and unfathomable amounts of data suddenly become digestible. Why is that?
The difference a visualization makes
When I look at a table I can remember a few data points at a time, but certainly not all of them. Numbers isolated from their broader contexts are hard to make meaning out of.
Graphs, on the other hand, give us the gift of seeing a representation of all the values at once, where we can easily and rapidly compare them. They take us out of the deluge of details so we can understand the bigger picture.
To see this in action, look at the data below about revenue in the movie industry. Spend some time examining the default table visualization. Then, try out different visualizations by clicking on the white icons in the Visualization tab and experience the effect for yourself.
The ability to see the big picture in this way illuminates a dataset’s essential characteristics. When we’re aware of the general landscape, identifying outliers is easy. Aberrations can be seen and addressed efficiently.
An easily visible outlier
It’s not all about outliers, though...
Data visualization helps you make connections
Relationships across data points matter—that’s where patterns emerge. We can draw connections between data points in a visualization by paying attention to characteristics like their relative areas, lengths, and positions.
Communicating these connections can effect significant change. Such was the story Jon Snow, who used data visualization to draw attention to a major issue of his time.
In addition to his remarkable fictitious life in Westeros, Jon Snow lived in London in the mid-1800s. He created one of the most famous examples of a data map ever known. This is an adaptation of it:
Courtesy of Robin Wilson
This is a visualization of the severe cholera outbreak that rocked London in 1854. Upon reviewing hospital records and talking to the sick, Snow, a physician, began to suspect that an infected water pump on Broad St. was to blame.
This went against the accepted theory of the day that disease spread through “bad air,” so he needed proof.
Snow created a dot map to illustrate the proximity of cases (red) in relation to the water pumps (blue) and convinced officials to remove the Broad St. pump handle. The outbreak quickly stopped.
This is a perfect example of the power of visualization to save lives. If you know of other noteworthy stories, let me know, and I might feature them in a future blog post.
Data visualization helps you communicate clearly
Jon Snow’s map solidified his suspicion of the Broad St. pump as the source of disease. It also helped him communicate that finding clearly to the people who could do something about it.
Even if you aren’t pioneering the germ theory of disease, you can still harness visualizations to tell stories with data and help others see what you see. To learn more, check out our eBook, “The Art of Telling Stories with Data.”
Let’s put this storytelling idea into practice.
Imagine you work in city planning, and you need to determine whether planning for more apartment buildings or more single family homes makes sense for your municipality. You look at this data table showing homeownership and rental rates in the US over time. A trend is discernible, but it’s not that clear or compelling—it doesn’t make a great story.
By contrast, here’s the same data as a column chart:
Looking at this visualization, it’s obvious that 2008 saw rental rates increase sharply while homeownership rates fell. It’s also easy to understand how the impact of the financial crisis has stretched itself over time, causing homeownership rates to drop and the percentage of renters to rise. Based on this visual, you could make a sound case for planning more apartment buildings.
This visualization that ranked health supplements based on their popularity and evidence of their effectiveness for particular conditions demonstrates another practical application of data. With this information, you could communicate with your family why you’d rather take Vitamin D on a regular basis instead of Vitamin E, which has been shown to be dangerous in large amounts.
You can even use visualization to justify moving to a different state to pursue a career as a professional dancer, based on and the concentration of jobs available in the field; or simply to add color to your daydreams about that life (hello, Nevada).
The potential applications of communicating through visualizations feel endless. For more fascinating examples of how visualization can help us communicate, check out David McCandless’s TED Talk on “The Beauty of Data Visualization.”
To recap, visualization matters a lot because it can help you:
- Quickly make sense of overwhelming amounts of information
- Draw connections with far-reaching implications, as in the example of Jon Snow
- Communicate clearly and tell stories
Good visualization is so important that we decided to dedicate a blog series to it. Look out for upcoming posts on viz-centric topics such as:
- How to choose the best graph or chart for your data
- Designing dashboards for UX/UI
- The problem with pie charts
Until next time,
Content Strategist, Customer Education