By Takin Babaei-Oracle on Jan 19, 2016
By Barry Mostert
“All I see are the same bar charts, pie charts and scatter plots that I’ve always had.” “I don’t see much difference from what’s in my current dashboards.” “Where are the new, innovative, different types of charts?”
When I talk to people about data visualization, initially they express disappointment that they’re not seeing “new ways” to present data. Actually, data visualization isn’t about replacing reliable forms of graphic expression; after all, a pie chart will always be pretty good at easily portraying a metric against a single dimension. A visual’s intent is to present information that allows the viewer to quickly digest the meaning of the data and in a way that will help to lead them to new insights. My experience has shown me that new ways of presenting and enhancing graphics are often poorly applied – and often for novelty effect. Adding effects like shadows, 3D, or making them spin doesn’t actually improve the information being represented. Sure, they may be “prettier” or “more fun” and so they’re great when doing a presentation, but these effects do little to support analysis of the underlying information – actually, they are more likely to create a distraction than to add value.
Of course, with data visualization tools, brand new types of charts are available, and I will discuss new chart types and the use of the technology called D3 in a subsequent post. For now, keep in mind that this isn’t the major benefit of data visualization.
“So if data visualization tools aren’t to provide a broader array of graphics, why is there so much hype about them and why shouldn’t I just stick with my classic dashboard tool?” The answer lies in the way that users interact with these two tools. A classic dashboard is usually created an answer to a specific, predefined question whose answer changes over time (e.g., revenue to date, production throughput). Classic dashboards are usually built to help users discover what has happened. Data visualization techniques, on the other hand, empower data exploration by supporting a guided conversation through the data. There is no predefined requirement and no predetermined end point - the conversation may lead anywhere. Data visualization goes beyond what has happened to encourage the user to uncover why it is happening.
The difference between data visualization and a classic dashboard is not in the appearance, but in the type of question that initiates the conversation with the data, and how you employ the visuals to answer the question.
In subsequent posts, we will offer more thoughts and observations about data visualization – applications, techniques, considerations for people, and tools.