My third assessment of business intelligence focuses on upcoming technologies that may or may not have a significant positive impact on effective use of BI. It is done in the spirit of one of President Lyndon Johnson’s most meaningless quotes, from a speech he gave in the 1960s: "My fellow Americans, the future lies ahead."
The point here is that, as past business intelligence prognostications have shown, predictions are as likely to be wrong as right. We can, though, be sure that things will change in one way or another; the future cannot be avoided. Therefore, these technologies are worth kicking the tires on to look for possible benefit, even if not all of them pan out.
Dynamic Data Visualization
The first is "dynamic data visualization." As we have seen, the search for ever deeper insights in business intelligence has led us to ad-hoc and rapid data analysis. However, this analysis is one-dimensional and typically does not consider a customer or prospect over time. Providing visual representations of the customer’s characteristics at a snapshot in time (usually now) is well understood; seeing how the customer’s needs are changing over time -- with fads, due to the customer lifecycle and for other reasons -- is not typically displayed well to the analyst.
However, the first signs of such dynamic data visualization are beginning to appear, in products such as JReports and in the use of cycled radar to show weather trends on websites. These visualizations can be multi-dimensional -- for example, showing changes in the amount of money spent by a particular market segment by unrolling lines as well as the number of customers involved by balloons at each quarterly report.
Using these and other tools, therefore, BI users can begin to project how customers will change their buying habits individually and en masse in the future, based on how they have been changing in the recent past. In turn, that means a more agile enterprise can anticipate customer needs and desires instead of reacting to problems with that customer, creating a far better customer relationship.
The second new technology, by now well-publicized, is semi-automated knowledge-based analytics, of which IBM’s Watson is the outstanding example. There are notable caveats with this approach, including the fact that the confirmation bias that leads a company's business intelligence team to overlook what they didn’t see now shows up in the knowledge-based tool, where it represents what the vendor expects that company BI wants to see.
However, in the areas where it presently is applied (such as biotech) and probably many others, semi-automated knowledge-based analytics is allowing users to cover a lot more ground and even do some ad-hoc "serendipity" searching for new compounds.
The third new technology is the sensor-driven Web. It is now clear that enormous amounts of current data, from tweets to GPS scans, flows through the Web and can be dipped into, like a net cast into a stream. What streaming databases do with data arriving at the enterprise can also be done with the typically Hadoop-file-system-stored sensor data that flows across the Web, and can also be filtered to zero in on customers and prospects.
This can occur not just in social media sites, but also in consumers' day-to-day actions outside the scope of the firm’s typical ability to see the customer. So enterprises, with some effort, can piece together solutions that capture and satisfy a much more multi-dimensional "customer of one" by using analytics on sensor-driven Web data.
As I noted above, these are technologies that I believe are worth watching, for significant benefits that can be gained today in some cases, and also for their potential for future benefits across a much wider scope of markets.
There still remains the question: What should we be aiming for in the long run as business intelligence evolves? What’s the endgame?
Wayne Kernochan is the president of Infostructure Associates, an affiliate of Valley View Ventures that aims to identify ways for businesses to leverage information for innovation and competitive advantage. An IT industry analyst for 22 years, he has focused on analytics, databases, development tools and middleware, and ways to measure their effectiveness, such as TCO, ROI and agility measures. He has worked for respected firms such as Yankee Group, Aberdeen Group and Illuminata, and has helped craft marketing strategies based on competitive intelligence for vendors ranging from Progress Software to IBM.