InetSoft Webinar: Business Intelligence Has Remained the Number One IT Priority

This is the continuation of the transcript of a Webinar hosted by InetSoft on the topic of "How Data Discovery Software Uncovers Stories in the Data." The speaker is Abhishek Gupta, product manager at InetSoft.

And last introductory comment to make, I noticed a high level theme from the Garter business intelligence summit. Business intelligence has remained the number one IT priority. They believe that the penetration of end users will increase 2.5 times this year. That’s a huge increase.

Why? Because of four key driving technologies: in-memory-data-management which we are going to talk about today, data visualization which we’ll talk about today, social software which tracks interaction between people, which we are not going talk about today, and search.

There are promising new models that are emerging. One is called the information buffet. Structured decision making from the top goes to more autonomous decision making spread out throughout an organization. You go from controlled and qualified access to open and equal, or open, unqualified access.

This is the concept of information democracy, and we believe this is fundamental for business decision making and the continued growth and penetration of business intelligence.

Sub Pieces of Business Intelligence Model

So now let’s look at the sub pieces of this business intelligence model. The first is in-memory data management. I’m going to quickly get to an example here. The idea here is to support the concept of the end users’ analysis by enabling very fast interaction, slicing and dicing and calculation. No predetermined data structures are required. The analysis can be completely ad hoc against any combination of any elements of any table in the memory pool.

And you will see in the coming example this could be 70 tables. It could be 10 tables, one table. It can be a lot of data in-memory. And since the detail is all in-memory, as the end-user slices and dice the data, the detail list is constantly changing and then when done, the end-user can easily export his or her list of customers, products, underperforming employees or whatever to another system for action.

In summary, the advantage of fast and easy analysis is that it can be ad hoc. It’s very fast. It’s really flexible. It’s great because you can work across tables and data sources. It's dynamic and can do that on the fly versus requiring pre structured data models. It’s very simple query engine especially with the strong visual front-end, and it allows you to get from the high level summary right down to the detail.

Usually it's the detail that is needed for tactical action to get taken. For example, it's great to find a case in your network where flights are really delayed, but at some point you actually have to do something about the specific planes or flights. Replaced them, or put new equipment on them. So somebody has to get to that level to take action.

“Flexible product with great training and support. The product has been very useful for quickly creating dashboards and data views. Support and training has always been available to us and quick to respond. - George R, Information Technology Specialist at Sonepar USA

Applying BI to a Higher Education Fundraising Call Center

So as we move into the demo, we are going to look at a call center system where there is a set of data tables. There are call records which are coming from the call system pictured by these blue cubes. There is alumni data. This is a higher education fundraising call center with a set of tables represented by the green cubes.

There might be e-mail appeals going out to the same people represented by the orange cubes coming from a marketing system. They use students for callers, and all of the student information is in another system, the registration enrollment system. So these are additional tables represented in red, in different systems which are loosely connected typically.

The first step is we mash those tables up with a drag and drop query builder. Then we build an interactive visualization which brings all of that data into memory. It could be as in this case 50 or 70 tables. We mash them together in-memory. We leave them alone, but then when a selection is made, it propagates through this set of charts. To get the answer to the question, for example, who is not doing well calling my top donors, you would need to get some information from some of the call tables, some of the alumni table to look for patterns and trends.