I've been scratching my head over the evolution of BI-as-we-know-it and the intersections of that with the increasing flow of information services in the channel and at large on the Web.
It's no surprise that people are looking at more and more internal and external resources for informed decision-making. In the internal case, data integration is a foundation of master data management as well. But integration for BI to common visual tools is increasingly taking place in subsystems and applications as well as relational databases and cubes - and in the visualization layer itself. Web services and SOA are moving that model too, though Darren Cunningham of Informatica reminded me on DM Radio last week of the importance of agnostic integration software interfaces to multi-siloed data pursuits.
The Web is another animal and last time I was talking about how much service-based information was flowing through application programming interfaces, and really taking over the traffic compared to the traditional Web siteor browser for that matter. To me, this is a leading indicator of the growing importance of service-based data/information feeds.
Only an hour passed between the time I posted that note and when I got an email from Omniture, one of the big shots of Web analytics, informing me that they too were about to release a product to monitor API traffic on the Web, and a lot of keyword tracking to measure application traffic and consumption patterns.
It was another reminder that it’s the information, not the presentation layer, that is going to matter in the future. "Essentially you’re virtualizing the information that consumers want by exposing it to different interfaces, different applications, different channels people want to consume it through," Matt Langie at Omniture told me. "So no longer are you requiring an end user to go to my site and look at my map there, I can now look at your map through another browser, through a desktop widget, through an iPhone app."
That's important because measuring traffic flows and keywords tells companies where to put their money and other assets. For example, BestBuy, an Omniture customer, is looking at their API traffic to third-party resellers to see whether their partners' choice of presenting pictures first or prices first sells more product.
Flip that idea around and API and keyword traffic can also be used to develop feeds of comparative intelligence for consumption through an BI analytic or visualization tool, and there seems to be a good bit of interest among vendors to accommodate all kinds of data feeds.
I'd interviewed Marge Breya, who heads up SAP's Intelligence Platform Group, for a chat about Business Objects Explorer for an upcoming issue. Explorer is a kind of intuitive front end that looks at end-user activity among related topics and pulls what it knows from related data feeds. Part of the goal is to eliminate the search box by calling up all the reports, inside experts and outside resources to follow what it assumes the topic to be. With an in-memory database, SAP's BW Accelerator indexes and presents the information to Explorer, the newest version of which is agnostic to data feeds. Excel or Teradata or Google, it doesn't matter.
Breya says the horizon includes technologies SAP calls Live Enterprise that will gather any kind of automated data source coming in. "You want to look at whatever kind of signal you want to make real-time decisions off of, whether it's, voice, video, data or anything else," she says. "We absolutely see this as a mashup world where we need to be putting together structured and unstructured content, mapping things back and forth so there is context around it."
I have seen enough "intuitive" interface engines to know machine logic isn't yet up to its human counterpart, though I was impressed by a demonstration of Explorer that progressively aggregated the data it offered around a theme.
On a separate note, context isn't the same thing as integration, so I imagine we'll be looking at apples and oranges as we ease the burden of gathering topic-related information assets. A data warehouse may be the last, best word on database management and analysis, but it's no longer the only game in town.
Many more vendors are going to chime in on this topic with marketing and products that will be better or worse. And if there's yet no roadmap to the future of analytical data consumption, contextand/ and integration, at least there's a road.