InetSoft Webinar: Key Points of Style Intelligence

This is a copy of the transcipt of a Webinar hosted by InetSoft. The speaker is Mark Flaherty, CMO at InetSoft, and he discusses the topic of “Embedded Business Intelligence”

Mark Flaherty: We believe that if you want to deliver these breakthrough applications, you need to a software platform that can give you three things. The first is the ability to store information and that information has to be more than just rows and columns, it has to be unstructured data, video all other kinds of things and we think that if you can do that in a unified way, it cuts your costs and makes the other parts of this a lot easier.

The second thing we believe, the connectivity is vital. And connectivity doesn’t just mean application integration, no that’s the big part of it. Connectivity is also about how you connect to users and how you connect to communities whether there are other vendors or are there people surrounding your institution, you need the tools and services to be that.

And third, insight is the ability to use the software to look at the information that’s locked in there and deliver real analytics to users at the point they are using it. So from inter-systems business point of view, we take those three features and we attach product names to them.

But a key thing as this diagram tries to show is that throughout our platform is one entity, you can install this whole thing in about five minutes, and you get everything I am talking about here. So we haven't taken, assemble a whole bunch of pieces together and hope they work. We have really started from the idea that all of these capabilities need to be woven right into the offering, we give to developers.

And underpinning all of this is our cache multidimensional database and that is an extremely flexible and reliable data storage engine that lets us store not only relational tables and OLAP data, but also has some very powerful ways of dealing with XML and some very interesting ways of dealing with unstructured data, including the ability to kind of make sense of that unstructured data without having to go through all the effort of pre-building ontologies. So we can load things like clinical notes or case notes or loan officer notes and we can infer a lot of knowledge from that without knowing a lot of things upfront.

We have a very powerful data integration product that gives us through either web services or variety of other means, a complete way to connect to other applications. It also has built in complete Web 2.0 framework for letting us build very quickly web applications without having to add a lot more layers of middleware. And most important to this discussion is we have an analytic engine in Style Intelligence and we are going to pick a little bit closer look at it. But Style Intelligence works very closely with both the underlying data storage and with the connectivity because we are going to see that is how we think that you can deliver on this promise of not only embedding BI, but allowing the users to pick an advantage of the business intelligence to drive other actions.

So let's take just a really brief look at what we think are the key points of Style Intelligence. This is not intended to be a product demonstration, where we are just really capturing the high level view here. And there are really three parts I think that are very important when we talk about Style Intelligence. The first is that we try to deliver data in the time that you need it.

Earlier here, we saw that there is a lot of pressure on people to deliver analytics closer and closer to the point of view of one of the things really happening. So we have built a lot of technology in Style Intelligence so that we can build models and incrementally update them. So if you build an analytics model, you can actually feed data into that from your transactional system. We are smart about updating all of our aggregates and indexes and all those structures, so that as your data changes, all of your dashboard inquiries change in real time as opposed to the more traditional, ‘let’s build the data warehouse and wait for the data to be built’. We talked about build-in ETL because our fundamental platform is a robust object oriented system. We actually have a complete object oriented environment for the modeling of your data model, but also for automatically generating all the transformation that needs to be done. So we can infer from your model what transformations need to be done and we kind of eliminate that whole ETL to data warehouse stuff that lot of other products do.