Data visualization has been an issue ever since the development of screens. I thought it might be interesting to look at how business intelligence (BI) has changed over time. On the video I use SAS Visual Analytics (VA) to show you.
Since the best examples are always practical ones, and I also wanted to look at something historical, I downloaded freely-available demographic information about Vienna.
The first stage in the evolution of reporting is cross-tabulation. It is probably not an exaggeration to say that without the cross-tab, there would be no reporting, no Excel, probably not even multidimensional databases. This means that no BI or analytics software can do without them even today. Why? Because the cross-tab has one absolutely unbeatable feature: everyone can read them, intuitively, and without instructions.
You can see all the totals and the details at a glance, which is the great strength of cross-tabs. But that strength is also their greatest weakness. They include everything, without exception and that makes it difficult to find trends or outliers
Fortunately, software developers have also spotted the problem, and provided functionalities for “tuning” the crosstabs, and making them easier to read. On the video you can see what happens when adding an interactive filter option and with some scaling. And there is also conditional formatting, another technique to get more out of a cross-tab.
The result is amazing. The report starts to tell a story. It is possible to create comprehensive reports from what started out as a series of numbers, just using relatively simple techniques.
By this point, however, we have largely exhausted the possibilities of cross-tabs. To get better results and more insights, we need other options.
A picture tells 1,000 words
Cross-tabs are very good at showing the detail, but it is much harder to see the overall picture, or get a feel for trends and distributions. There are some techniques that can help, but overall, you still have a table of numbers.
It is time to bring in modern techniques, and particularly use of pictures.
Charts are particularly good at showing information about general trends and distributions. The drawback is that you lose a lot of the detail.
But unlike the graphs you drew at school, on a computer this is no longer a problem. The underlying data are still there, and you can look at the details at any time by linking objects. The only problem is that you can only look at the detail in the portion on which you are currently focused. So you do not have all the data at the same time as you do with a crosstab.
You can, however, embed small charts into lists or crosstabs to make a particular point. We call these micro-charts.
Maps and, of course, their keys, have changed a lot in recent years, and the potential is now huge, even though not all data are suitable for this type of presentation. As with bar charts, you have the same trade-off between overview and detailed information, whether you use flexible Geo-bubble charts (above) or Geo-Region maps.
But perhaps the biggest change in recent years in visualization tools is animation. This enables us to move from reporting through to analytics.
Clever as this is, more is possible.
Moving into the future
I think there is much more to come from data visualization. Instruments such as forecasts, scenario analysis, decision trees, and text analysis offer many more options. The journey is only just beginning.