Friday, October 21, 2005


I have issued a number of probes to BI vendors for views on the potential synergies between SOA and BI - what I've been calling Service Oriented Business Intelligence (SOBI).

Service-Oriented Business Intelligence (SOAPbox blog)
Service-Oriented Business Intelligence 2 (SOAPbox blog)
Web Services for BI (CBDI Journal, June 2003)
Service-Oriented Business Bntelligence (CBDI Journal, October 2005)

This week I had a useful briefing from Information Builders, makers of WebFocus. I first encountered Focus when I was working with Fourth Generation Languages (4GL) over twenty years ago, and it's interesting to see how the present stance of Focus (which I must now remember to call WebFocus) represents both continuity and change.

Information Builders certainly seem clued up about web services and SOA, and claim to have been playing in this space rather longer than some other BI vendors.


IB's main entry into the SOA space is a product called iWay, which calls itself an "Adaptive Framework for SOA" and claims to be "a complete toolset for creating composite applications and reusing existing IT assets". iWay is marketed by a separate company, iWay software, which is a wholly owned subsidiary of Information Builders.

One interesting piece of functionality is that it can be used to make data from legacy systems (such as unstructured data from notes fields) visible to the Google Enterprise Server, and therefore available for aggregation and analysis. May be worth a look.

Value of BI

4GLs were always supposed to improve productivity for developers, and to improve access for end-users. This twin agenda clearly remains in force. Information Builders sees business intelligence as a key source of value to any business organization; so the greater the number of people using the BI tools, the greater the potential value to the business. (Well, they would say that wouldn't they?) There is therefore considerable emphasis on improving the accessibility of BI functionality, as well as achieving economics of scale in the delivery of BI functionality.

Of course, technology vendors are naturally prone to make optimistic statements about the value of their tools, and the importance of having everyone using them. In this case, realistic assessments of the value of BI must depend on analysing the potential to improve business processes. Information needs are integrated with specific business responsibilities. Business processes and services may be improved by introducing effective feedback loops - for example if you allow customers to access restaurant inspection data, the dirty restaurants disappear pretty quickly. Note that these control loops typically go outside the boundaries of a single organization.

Instead of business intelligence being a highly specialized function, restricted to head office wonks with expensive and complicated gear, the power of business intelligence is taken to the edge of the organization - together with the corresponding accountability.

Integrated BI

I agree that the business value of BI may be greatly enhanced when BI is integrated with the business process. I have been calling this integrated BI; Information Builders talks about operational BI or pervasive BI, which are perhaps not quite the same thing, but are at least broadly in the same area.

There are lots of integration techniques that are relevant here - not just web services, but also Web 2.0 technologies such as Atom and RSS. For example, it is possible to subscribe to a complex enquiry. But we can push this further - imagine being able to subscribe to a hypothesis, and then being notified whenever any relevant evidence (for or against the hypothesis) becomes available.

Collaborative BI

I have argued that the next step beyond integrated BI is collaborative BI - supporting collaboration between distributed knowledge workers. Having implemented simple versions of Embedded BI and Integrated BI, Information Builders has announced the intention to introduce some support for Collaborative BI in the 2006 release of WebFocus. I look forward to seeing the details of this. However, I suspect that the full power of Collaborative BI will take longer to develop.

BI Process

So how is all this technological product innovation going to be reflected by process innovation - affecting the way that people build and use BI systems and services? Perhaps it is too early to say. Vendors like Information Builders may contribute to this innovation, may disseminate patterns and best practices, and may wish to develop formal methodologies. But I suspect the important changes will emanate from the user community, and will be slower to emerge.

No comments: