Eric Kavanagh: Okay folks, welcome back here to DM Radio. We had a very good opening segment with our friends with Deloitte and Third Nature, and they will jump in on some of these following segments as well. Next up we have Byron Igoe from InetSoft. Welcome back to DM Radio.
Byron Igoe: Thanks Eric, good to be back.
Eric Kavanagh: Sure thing. So, in our quest to come up with the top 10 reasons to use data visualization, what are couple of reasons that you have got?
Byron Igoe: Well I mean I definitely think that the interactivity and big data implications, and all the things that have been talked about so far are very near and dear to my heart. But I was actually going to focus on just the core idea of visually displaying information. So I am just getting back to normal charts and graphs as opposed to traditional tables or cross tabs and things of that nature.
| #1 Ranking: Read how InetSoft was rated #1 for user adoption in G2's user survey-based index |
|
Read More |
So there were a couple of things that I had in mind. One, obviously there are some tasks that are just better suited for looking at the data in a visual manner, right. The first one that pops to mind is trending. If you are seeing data spread out over at a timeline, obviously you can see the shape of that trend much better in a visual display. I mean nobody is going to argue that you can't see an increase or a decrease by looking at the pure numbers but in terms of is this trend a linear increase, is it logarithmic or exponential, you can get a sense for that pattern a lot better when you are looking at a chart or graph.
Similarly, when you have got multiple variables that you are trying to analyze, I think that correlation between those variables is much easier to do when you are plotting it on a scatter rather looking at a couple of columns in a table. I don’t think that we as humans are very good at sort of on-the-fly looking at numbers and then computing the correlation coefficient between them.
And then I guess also identifying outliers, and this is very similar to the trends or correlation. When you are plotting multiple variables together using some of the more advanced displays that you were talking about before, rather than just the standard bars and pies, you can really see the exceptional pieces of information jump out at you, really exploiting the human eye’s ability and the brain’s ability to identify outliers and spot the odd one out amongst the pattern.
Eric Kavanagh: Yeah. And obviously that’s very useful stuff for identifying patterns, right, because the whole goal of using data visualization, the whole goal frankly of Information Management is to identify possibilities, to identify opportunities to get new revenue or to streamline or to do things that will improve your business. So it seems those outliers are a pretty critical part of the whole picture, right?
Byron Igoe: Exactly right. And that exactly brings me to my second point, perfect segue, into actual decision making. So analyzing the numbers is just one step of the process then you actually have to make a decision based on what you have analyzed. And so actually studies have been done. One case in point was the finding from a report out of St. Cloud University. Without going into all of the details, basically they asked a set of people to estimate when a company would be bankrupt given some financial information.
And so they gave this information to two groups of people. One set were finance people so subject matter experts. Another group were sort of the lay people, not subject matter experts. Obviously they were intelligent ,people but they weren’t trained in analyzing when a company goes bankrupt. And then they also presented the information to these two groups in two different ways either tabular, just the data, or visually in a chart.
And so they found that among the experts, there was virtually no difference in the accuracy of their estimation, whether they saw a chart or just a table of the raw numbers, experts were able to make the decisions equally well. However, there was a difference among the amateurs. So the amateurs were much better at estimating the time when the company would go bankrupt when they were looking at the visual display of the information rather than at the numbers.
And so, in companies, this means that if you are not an expert, it is going to be helpful to see that information visually, and furthermore, if you are an expert but you are trying to convince somebody else or persuade somebody else who is not, you are better off showing them a visual version of the information rather than the raw numbers.