Chances are that you have heard about big data, perhaps even worked with it, and even tried to make sense of it in a business context. And when it comes to business intelligence (BI) and relational databases, the ability to make realtime decisions is critical.
Now enter data visualisation and its ability to present the next step for more effective decision-making. But what is this term that is surfacing in the broader BI sector?
Says Martin Rennhackkamp, CIO of PBT Group: “To really understand the benefits of visualisation, just think how difficult it is to extract value out of spread sheets and tables when all you are looking at is a collection of numbers.
“Not only is this an incredibly cumbersome way of trying to determine patterns that would ultimately drive business strategy, but it often requires specialists to get meaningful value out of it. According to Vitaly Friedman, the main goal of data visualisation is its ability to visualise data, communicating information clearly and effectively.”
This boils down to providing users with data that is presented in such a way that they will be able to detect patterns between seemingly disparate information points. These patterns are very hard (some might even say impossible) to detect and analyse when the data is presented in a more traditional spread sheet format.
Continues Rennhackkamp; “For all its merits, visualisation is not just a case of converting a few tables into a graph and hoping that it will unlock the secret to the universe. World-renowned visualisation expert, Stephen Few, uses the term BI Visualisation when he says that data visualisation is more than just displaying the numbers in a graphical format.”
Already numerous tools such as The Data Visualisation Catalogue exist to help designers, economists, statisticians, and even scientists decide on the right aid for their specific requirements. If terms such as circle-packing, density plot, and nightingale rose chart sound even more complex than the traditional alternative, rest assured that they fulfil very specific needs.
“As with any technology or trend, it is often quite easy to start using it but it takes considerable skill to understand which types of graphs will suit the data you have. These graphs need to be selected based on their ability to highlight the points of interest and other issues that will enable you to make more effective decisions,” continues Rennhackkamp.
“Think about the rise of infographics in the past two years. You get those that provide amazingly detailed information at a glance and then there are those that simply do not make sense. The same holds true for Visual BI.”
A key aspect of visualisation is the concept of Data-Ink ratio. While this might not sound incredibly exciting, this concept was introduced by Edward Tufte in 1983 and boils down to only presenting the actual data in visual format. The reason for this is to avoid drawing the attention of users to the data representation of irrelevant elements.
After all, the best data visualisation in the world means nothing if users are not able to make sense of it. It might be so complex that very little value is extracted. And even if some patterns are identified, they might be potentially incorrect due to all the additional clutter.
Concludes Rennhackkamp: “It is with this in mind that I stress that you carefully consider your strategic approach when it comes to data visualisation. All the good intentions in the world will mean nothing if you do not understand what kind of graph you need to select for the kind of information you want to extract, for that information to be beneficial to your business.”
