We are living in a data intensive world today. It is estimated that we generate 2.5 quintillion bytes of data every day (that is about 2.5 billion GB, or roughly equivalent to a billion HD movies every day)! About 90% of all the data in the world now has been generated in the last two years, and this is going up on a daily basis. This makes it obvious why IT companies worldwide are investing in big data technologies and analytics solutions. These help identify trends and key insights from the massive amount of data being generated, which can help businesses shape their strategy. However, the output of these analytics solutions has to be presented to the user in a way that the key insight becomes easily discernible for it to have the desired effect. This is where effective data visualization becomes critical.
A picture is worth a thousand words. That sums up the essence of data visualization.
Companies have been visualizing their data for decades in their reports and presentations. But the important questions are – Do these visualizations serve the purpose? How many actually provide the insight to its viewers? How many reach the audiences in the manner they were intended to be? As highlighted earlier, the large amounts of information being generated can lead to so much data lying unattended and neglected. You can get overwhelmed by the amount and variety of data out there. This could create situations where one tries to cram a lot of data into a visual form, where the data points lack cohesion and the audience struggles to understand the data.
Data visualization is essentially a way of finding a solution, which helps map a way to the required information quickly and also derive more value from the data, which otherwise lies hidden behind the big charts, tables and text.
While still a novice, according to me good data comprises of three important aspects – a catchy title, a good information architecture, and a good visual representation of data.
I would love to expand this blog and include all that I want to say about data visualization, but it takes time to collect examples, so we will go one aspect and example at a time.
Defining a title:
Good data visualization always starts with an apt title. The title should be like a prologue, giving the viewer a sense of what to expect from the data below. While from the viewer’s perspective the title is the first thing that meets the eye, the title can be picked at the end based on the key insight that is being highlighted.
Often the title just describes what the data below is, but it will be more effective if the title gives the reader an idea of what insight is being conveyed. For example, in a report on the rural consumption story in India, a simple bar chart that shows rural household income could be titled in two ways – (1) Rural household incomes per month, 1951-2011 or (2) Rural household income – Tripled in last 60 years. While the first title is factually correct, it does little to convey what the chart below is about. The second title on the other hand tells the reader what exactly is being highlighted in the chart below. This makes it easier for a user to appreciate the data presented and helps it fit in with the context of the report.
Let’s look at how this gets better with another example.
While the title could’ve been as plain as “Barack Obama wins by x number of votes”, the visualization has been given a title which goes with the visual representation method, also expressing that the votes came in pieces from various regions across the nation to determine Obama the winner. A title here acts as the cherry on the cake.
The visualization above shows the cause of deaths over the years and how animals and insects too play a role here. There is an interesting title to the graph – which of course looks like mountains.
Keep watching this space for more interesting examples that we come across. We will discuss the other aspects of data visualization in my next blog.