Data visualization, a tool by which facts, statistics, or other complex pieces of information, are presented in a format where trends and developments are easily grasped and understood. Our eyes are naturally drawn to color and focal points. Visually-immersive presentations using graphs or charts can effectively pull attention to the key points of research and analysis. This becomes particularly important when it comes to decision making and responding to new events in a timely fashion. A spreadsheet or data script, while often a necessary part of information gathering, can also be confusing and overwhelming.
Even after key points are identified and understood, there typically remains the risk that changing or adding new items of data could once again cloud things for a period of time. For example, a store manager might not understand why sales were down a particular month, when no changes were made at a staffing or supply level. With data visualization, defining trends and patterns becomes much easier as they quite literally appear in front of us. This is one of the many reasons why data visualization is important: It’s “a quick easy way to convey concepts in a universal matter,” [i] where experimentation with the input can be done quickly, and the consequences of varying or experimenting with that input can be seen quickly as well.
Clarifying Problems with Data Visualization
Another reason is correctional in nature, in that it can clarify what areas of the data are problematic or need attention. A webcomic by Randall Munroe presents several thousand years of average CO2 levels throughout the world in an interesting, scrolling format. [ii] Merely looking at the numbers might not give the full story. Due to variations and shifts that appear consistently throughout the chart. However, at the bottom of the chart is when we can understand that the numbers present a marked change from the pattern up until that point.
Isolating Relevant Parts
Data visualization is also important because it can isolate those parts of a series that are relevant to specific questions or topics. [iii] For example, a trend such as customer behavior is a complex and seemingly-arbitrary collection of data points. If one were to gather together all the different factors that influence a shopper’s purchasing decisions, it would, by necessity, be presented as a grid of crisscrossing numbers and dates. While the causes behind some of the trends here would be somewhat straightforward (i.e., people stop buying hats and mittens when the weather turns warmer), determining the reasoning behind other trends is typically requires a multi-step approach.
Game of Thrones and Data Visualization
For example, the TV show Game of Thrones had increasingly higher and higher ratings every week of its second season in 2012, building up to the climactic episode “Blackwater” . . . which saw the ratings plunge [iv]. They went right back up the following week for the finale, but simply looking at the raw viewer numbers presents more questions than answers. The episode in question had very good user reviews, as did the episode before it. What happened? The answer can be found by again, looking at the data. This particular episode debuted on Memorial Day, a three-day weekend, and many viewers simply happened to be out of the house that evening. We know this by looking at flight and travel data for a typical Memorial Day weekend [v]. By observing that, year by year, travel-related activity noticeably ticks up across those dates.
Determining How and Where to Sell Goods and Services
Data visualization helps us determine how and where to sell goods and services, or in other words, “which products to place where.” [vi] Location is a key value in real estate, and the same holds true here: No one will buy your product if they don’t know how to find it, and no one will know how to find your product if you don’t know how you’re placing it. To this end, companies routinely collect sales data on different products[vii]. The relative importance of each metric is presented in a very straightforward way. Price is considered most strongly. Followed by the type of store the product is sold in, and then various physical descriptors of the product itself. In other words, the chart demonstrates that the factors of affordability and convenience seem to outweigh many other considerations. Which suggests that the big takeaway is: Make it easy for a customer to purchase something.
Finally, data visualization plays a reactive role for businesses, and helps them to determine what products might be in demand in the future. Many companies use visuals to help them make sense of these trends, and often make sense of time series forecasting, or “[using] a model to predict future values based on previously observed values.” [viii] Essentially, they look at previously occurring sales events over a long period of time, typically spanning several fiscal cycles. Then use those numbers to make predictions as to what sales of a specific product should look like as time goes on.As before, there are often several metrics that are useful to graph in this fashion, including item sold, number of sales, date sold, location sold, how many customers were in the store on that date overall, etc. However, knowing how much emphasis should be placed on each metric, and what sort of emphasis, is a question that data visualization is well suited at exploring. A typical visualization of these data points can be presented in an easy-to-understand graph form. Where the peaks and valleys of any given arrow chart generally stand out pretty clearly, and often coincide with memorable dates and events. Sales of green shirts might spike around mid-March. Fireworks sales in certain states boom over holiday weekends, but especially at the beginning of January and July.
To the Point . . .Moving away from holidays, a downturn in customer activity might appear a mystery. Until the data visualization shows that a rival store opened across the street on that particular month, and many potential customers may have considered exploring a new option during that time. There are many more reasons to use data visualization. Ultimately, they remain a very good tool. Most importantly, they don’t just answer a great deal of variable questions about your target market. They make it easier to understand who your target market is and how they behave. What are some ways you use data visualizations to explain concepts and ideas? Please share in the comment section below.
ABOUT USCATMEDIA is an award-winning Inc. 500 company based in Atlanta, Georgia. Founded in 1997, the company specializes in advertising, creative services, media production, program management, training, and human resource management. As a Women-Owned Small Business (WOSB), CATMEDIA provides world-class customer service and innovative solutions to government and commercial clients. Current CATMEDIA clients include Centers for Disease Control and Prevention (CDC), Federal Aviation Administration (FAA), and the Department of Veterans Affairs (VA).
Stay Connected with CATMEDIA:For more information, please visit CATMEDIA Follow us on social media!
- [i] “Data Visualization: What It Is and Why It Matters.” SAS.com. N.p., September 2019. Web
- [ii] Munroe, Randall. “Earth Temperature Timeline.” xkcd.com/1732. N.p., December 2016. Web
- [iii] “Data Visualization.” SAS.com. N.p., September 2019. Web
- [iv] Hibberd, James. “Game of Thrones Ratings Dip for ‘Blackwater’.” EW.com. Meredith Corporation. May 2012. Web
- [v] “Holiday Season Sales vs. Memorial Day Sales Trend” trends.edison.tech. N.P., May 2018. Web
- [vi] Vadapalli, Sricharan, Hands-on DevOps, Packt Publishing, Birmingham, UK, 2017, p. 31
- [vii] Olton, Andrew. “Data Science Case Study: Optimizing Product Placement in Retail.” towardsdatascience.com. N.p., May 2018. Web
- [viii] Li, Susan. “An End-to-End Project on Time Series Analysis and Forecasting with Python.” towardsdatascience.com. N.p., July 2018. Web