Karan Passey

Data Visualization 101: How, What, Why?

Data Visualization 101

A picture is worth a thousand words.” This old, English idiom could not ring more true than in today’s fast-paced, digital age – the big data age. At a time when we are creating 2.5 quintillion bytes (or 2.5 million terabytes) of data each day, executives and decision-makers across the globe are looking for ways to turn complex and voluminous data into comprehendible and comprehensive, actionable insights. Enter, data visualization.

What is Data Visualization?

The visualization of data for purposes of analysis is not a new concept. Finding their roots in Descartes’ Cartesian coordinate system, several graphical diagrams such as the line, area and bar chart were invented in the late 18th century by Scottish engineer and political economist, William Playfair. He was also the inventor of the once widely-popular, yet more recently denounced, pie chart.

Data Visualization sits atop the Big Data Analytics pyramid (Figure 1) and is often the only layer that is visible to executives and other decision-makers. Thus, the success or failure of a Big Data analytics program often depends on the success of the visualization layer. A company may have the most advanced data capture, storage, and transformation technology (and use the most complex algorithms and statistical models to analyze that data), but if the information isn’t displayed clearly, accurately and efficiently, the whole point of leveraging Big Data is lost.

(Figure 1 – Big Data Analytics Pyramid)

Why Data Visualization? What are the benefits?

It is often said that data is the new world currency. But let’s face it, raw data is boring and difficult to make sense of it in its natural form. Because of the way the human brain processes information, using charts or graphs to visualize large amounts of complex data is easier than poring over spreadsheets or reports. According to the Visual Teaching Alliance, studies show that 90% of information transmitted to the brain is visual, visuals are processed 60,000 times faster in the brain than text, and our eyes can register 36,000 visual messages per hour. It’s a no-brainer that visuals work better than text.

All types of organizations are using data visualization to help make sense of their data and to comprehend information quickly. Data visualization is a quick, easy way to convey concepts in a universal manner, and due to advances in data visualization technologies, you can experiment with different scenarios by making slight adjustments to available data filters – This is called visual analytics. Visual analytics allows users to directly interact with data, visualize relationships and patterns between operational and marketing activities, gain insight, draw conclusions and make better decisions, quicker. The visibility and clarity delivered by such digital technologies and advanced analytics can give executives unprecedented, granular views into operations. Additionally, it may increase agility and support better strategic decision making by showing the reasons why certain recommendations make the most sense. This can have a significant impact on how businesses gain insight from their data.

Data visualization ultimately aims to provide perspective, reveal trends, provide context and tell a story. Most importantly, data visualization should empower users to harness, in a meaningful way, the power of Big Data.

How to tell a story with Data Visualization?

One of the cardinal sins of a business presentation is for good data to be presented badly. How so? If data is misrepresented or presented ineffectively, key insights and understanding are lost, which hurts both the overall message and the reputation of the person or team delivering the message. Knowing your data and knowing how to best present it, are two ways of getting data visualization right.

Data comes in different types. The most common are:

  • Quantitative: information that can be measured and written down in numbers. For example, the outside temperature or the number of employees in an office.
  • Qualitative: information about qualities that cannot be measured. For example, the color of a briefcase. Most business problems deal with quantitative data. Purely qualitative data is very difficult to analyze due to the absence of numbers. However, qualitative data is powerful when used in a categorical context.
  • Categorical: information that can be organized into mutually exclusive categories. For example, six managers are from Asia, five from Africa, two from Australia, and nine from the Americas. In this case, categories are the regions in which the managers are located.
  • Geospatial: information that has a geographic component (in the form of coordinates, address, city, or ZIP code) to it. For example, locations of sales offices or well drilling sites.

Selecting the right type of data visualization involves understanding the underlying data and more importantly, the story you are trying to tell. There are four main presentation types that will aid you in telling your story:

  • Comparison: used to compare magnitudes of values among items, over time or both – For example, sales from different regions or sales of a particular region over time. Bar charts and table charts typically work best for comparisons among items, and line graphs and bar charts for comparisons over time.
    • Geographic Comparison: used to compare used to compare magnitudes of values among items, but displayed as a map. For example, the number of Fortune 500 companies by city within the United States.
  • Composition: used to compare part to whole relationships that are either static or compared over time. For example, percentage of online sales compared to other channels for a single period or over a number of periods. The pie chart has been a favorite for presenting static composition, however many experts agree that it is only best used when showing popular fractions (1/2, 1/3, 1/4). A Treemap or stacked bar chart may work much better. Stacked bar charts also convey composition information over time well.
  • Distribution: used to show how quantitative values are distributed over an axis (from lowest to highest) or into pre-defined categories. For example, number of customers per age group, further classified by income level group (second variable). A bar histogram or line histogram may work well for single-variable data, while a scatter plot and 3D area chart can work well for two and three-variable data, respectively.
  • Relationship: used to show the relationship between different data points of an item. Aggregating the relationship over multiple items reveals trends, correlations, clusters and outliers. For example, the correlation between marketing spend and product sales. Scatter plots and bubble-charts work best for two and three-variable data, respectively.

Data visualization is as much an art as it is science and numbers. In addition to using the right chart type to tell your story, you can also play with levers such as color, size, scale, shapes and labels to direct attention to the key messages of your story. The most important aspect, however, is getting the “story” right; ensuring that the right data is appropriately visualized for the audience that will consume it. Data visualization technology too, is constantly evolving. From automatically recognizing types of data to making recommendations about the best chart type to use, advances in data visualization technologies have made creating stunning visualizations a matter of a few clicks. More, the emergence and mainstreaming of virtual reality has given data visualization a new dimension, allowing for visualization concepts and techniques that have hitherto been restricted to futuristic sci-fi movies and TV shows. The future of data visualization is bright, and it will be exciting to see what advances and discoveries are made in the next few years.

By Karan Passey, Manager, Data & Analytics Enaxis Consulting

Cloud Syndicate

The 'Cloud Syndicate' is a mix of short term guest contributors, curated resources and syndication partners covering a variety of interesting technology related topics.

Contact us for syndication details on how to connect your technology article or news feed to our syndication network.

IBM Puts Data To Work for AI and in the Cloud

IBM Puts Data To Work for AI and in the Cloud

Artificial Intelligence, deep learning and cognitive solutions are a big part of the future of IBM, as clients accelerate their digital transformations fueled by data. The company’s new multicloud data platform, IBM Cloud Private (ICP) ...
Maduro Attempted Assassination by Drones - A Nasty Part of the Future Has Arrived

Maduro Attempted Assassination by Drones – A Nasty Part of the Future Has Arrived

January: I wrote of Slaughterbots – the potential of small ariel drones that could kill. March: I talked about how Sci-Fi was becoming reality with the intro of Skydio a drone that could track you ...
Analytics – A journey to AI

Analytics – A journey to AI

Artificial intelligence (AI) has been around in concept since the 1950s when Arthur L. Samuel, created a learning algorithm that allowed a machine to beat the local state checkers champion.  Yet, it took the largest supercomputer ...
Part 2 - Identity Assurance by Our Own Volition and Memory

Part 2 – Identity Assurance by Our Own Volition and Memory

Identity Assurance by Our Own Volition and Memory We believe that the reliable identity assurance (See part 1) must be built on three prerequisite principles as follows. Volition of the User – with Self-Determination, Identity ...
Four interesting Amazon patents in 2018 that use machine learning, AR, and robotics

Four interesting Amazon patents in 2018 that use machine learning, AR, and robotics

“There are two kinds of companies, those that work to try to charge more and those that work to charge less. We will be the second.”— Jeff Bezos, CEO Amazon When Jeff Bezos launched Amazon.com ...
SAP S/4HANA Cloud Adds Exciting New Capabilities to Manufacturing and Professional Services Industries

SAP S/4HANA Cloud Adds Exciting New Capabilities to Manufacturing and Professional Services Industries

Earlier this year, customers were provided with the deepest, broadest AI-powered functionality in the ERP market with the 1805 release of SAP S/4HANA Cloud, SAP’s flagship intelligent ERP solution. This unlocked 12 new AI-powered scenarios, which significantly ...
Top 5 Deep Learning Architectures

Top 5 Deep Learning Architectures

If you are a deep learning practitioner or someone who wants to get into the world of deep learning, you might be well acquainted with neural networks already. Neural networks, inspired by biological neural networks, are pretty useful ...
DIY AI! Build your own AI with kits from Google

DIY AI! Build your own AI with kits from Google

Worried about the rise of the machines? Like doing things yourself? Now you can address both. Build your own AI with kits from Google. You can tell when something’s time has come. Hobbyists and tinkerers ...
AI Meets Industrial IOT: Power Your Data-Driven Enterprise

AI Meets Industrial IOT: Power Your Data-Driven Enterprise

The rapidly progressing Artificial Intelligence (AI) technology offers a lot of new opportunities to businesses in the industrial sector. Smart machines that are capable of performing repetitive tasks with accuracy and self-correcting errors would be ...
Strategies for Monetizing Data: 2018 and Beyond

Strategies for Monetizing Data: 2018 and Beyond

Strategies for Monetizing Data The data revolution is here, and it creates an investment priority for enterprises to stay competitive and drive new opportunities. One of the brightest areas is data monetization, which describes how ...
Driving AI Revolution with Pre-built Analytic Modules

Driving AI Revolution with Pre-built Analytic Modules

What is the Intelligence Revolution equivalent to the 1/4” bolt?   I asked this question in the blog “How History Can Prepare Us for Upcoming AI Revolution?” when trying to understand what history can teach us ...
Artificial Intelligence And The Future of Accounting

Artificial Intelligence And The Future of Accounting

The Future of AI Accounting Artificial intelligence has become an extremely hot topic over the last couple years. While many people welcome the thought of self-driving cars and chatbot customer service agents, others are rife ...
The Future of AI is Here

The Future of AI is Here

For the last five years, we’ve been discussing the future of AI with many organizations, including Intel, at the forefront of that conversation. However, while ensuring the world gets to experience new heights and unparalleled ...
That’s It? After All the Hype and Fear About A.I.? Turns Out They Can’t Do the Hard Stuff

That’s It? After All the Hype and Fear About A.I.? Turns Out They Can’t Do the Hard Stuff

Well, guess we should have known better. A.I. can play Jeopardy, Chess and Go but it seems their capabilities have topped out. Let’s talk about a new A.I. “winter”. A.I. was going to take your ...
The Cloud Reveals a Future of Work That May Be Scary to Some

The Cloud Reveals a Future of Work That May Be Scary to Some

The Future of Work May Be Scary to Some Anyone paying attention to the world of work over the past 10 years or so has noticed the profound changes. Whether client-facing or internal, the cloud ...