Friday, November 19, 2010

Embedding Intelligence into the Business Process

Is the business process evolving from bureaucratic workflow towards some form of flexible intelligence? Some of us have been predicting this for a few years now, but there are some hopeful signs that it may finally be starting to happen.

In this post, I'm going to talk about two specific aspects of this.
  • Embedding business intelligence (BI) into the business process.
  • Embedding Enterprise 2.0 into the business process.

Embedded BI

The idea of embedded business intelligence has been around for many years. See my blog on Service-Oriented Business Intelligence from September 2005. See also my slideshare presentation.

When software vendors talk about embedded BI, they often mean embedding BI functionality in other pieces of software - for example ERP applications - to allow these applications to produce more interesting reports. There are several niche BI producers in this space, including Jaspersoft, Pentaho and Yellowfin. Brian Gentile of Jaspersoft talks about this in his recent article The BI Revolution: Business Intelligence's Future. TDWI November 10, 2010. For an article explaining the difference between Embedded BI and Integrated BI, see Execution MIH.

For BI to be embedded in the business process, we need to have an understanding of the business process that includes some cognitive task, such as a complex decision, where some business intelligence capability can be used specifically to support this cognitive task. In some cases, the aim might be to make the process faster and more efficient, but more usually the aim is to make the process more powerful and effective.

Embedded BI in this sense can also be related to intelligent event processing, where analytic capability embedded in one process can trigger automatic as well as human responses in other processes. Brian Gentle talks about this in an earlier article The BI Revolution: A New Generation of Analytic Applications. TDWI October 20 2010. 

Beyond embedding BI in the business process, we might look forward to a state in which analytics is embedded in the entire enterprise, what Tom Davenport and his colleagues call the Analytic Organization. (See my review of Competing on Analytics.) This is the proper meaning of the term Pervasive BI, which Dave Mittereder defined in 2005 as "empowering everyone in the organization, at all levels, with analytics, alerts and feedback mechanisms" (Pervasive Business Intelligence: Enhancing Key Performance Indicators Information Management Magazine, April 2005).

Embedded Enterprise 2.0

In her piece Time For Enterprise 2.0 To Get Enterprisey, Sandy Kemsley takes a sceptical look at the extent to which Enterprise 2.0 is supporting the core business.
"You hear great stories about social software being used to strengthen weak ties through internal social networking, or fostering social production by using a wiki for project documents, but many less stories about using social software to actually run the essential business processes."
She quotes Andrew McAfee:
[The CIOs] weren’t too worried that their people would use the tools to waste time or goof off. In fact, quite the opposite; they were concerned that the busy knowledge workers within their companies might not have enough time to participate.
And comments:
The fact that the knowledge workers had a choice of whether to participate tells me that the use of social business software is still somewhat discretionary in these companies, that is, it’s not running the core business operations; if it were, there wouldn’t be a question of participation.

It seems to me that there are two possible interpretations of McAfee's remark. Sandy's interpretation is that busy knowledge workers simply don't find time to do any Enterprise 2.0 stuff at all, and she concludes that if the business process can still work without it, then the Enterprise 2.0 stuff is discretionary. An alternative interpretation might be that the business knowledge workers don't have enough time to do enough Enterprise 2.0 stuff to get as much intelligence (requisite variety) into the business process as the business really needs. (I happen to prefer the second interpretation, but I don't know whether this is what McAfee really meant.) In other words, it could be that there a trickle of benefit rather than a decent flow.

I'm presuming that the way Enterprise 2.0 is used within the business process is to support specific cognitive tasks, such as interpreting and making sense of events, and making complex decisions. These tasks may be done by an individual knowledge worker, possibly drawing on knowledge made available by co-workers, or may be done collectively by a network of knowledge workers. The quality of sense-making and decision-making doesn't necessarily increase just because you have more people spending more time on it, but with highly complex business situations the opposite is almost certainly true - the quality will be impaired if you have too few people devoting insufficient time and attention.

But I worry a bit when technology vendors merely invoke the magic words "business process" without demonstrating any real understanding. For example, Sandy's blog links to Klint Finley's piece on Tying Enterprise 2.0 to Business Processes, or Creating New Processes for the Social Enterprise, which doesn't say anything I can recognize as being about business process; as far as I can see, it is largely about activity stream filtering as a technical solution for integrating pieces of software. Finley's piece links in turn to a Monday Musing by R "Ray" Wang which states that 

Organizations seeking a marketing edge must digest, interpret, and asses (sic) large volumes of meta data from sources such as Facebook Open Graph.

I think there may possibly be a business process implicit in there somewhere, but exactly how this business process would be supported by Enterprise 2.0 is left to the imagination. I hope "asses" isn't a Freudian slip.

To be clear, I can see how the technologies Klint and "Ray" are talking about might possibly be embedded into a sociotechnical system to support a real business process. But they aren't actually making the connection, nor are they providing any evidence that anyone else is doing so. Even Michael Idinopulos, who at least sounds as if he knows what he is talking about in The End of the Culture 2.0 Crusade? fails to provide any concrete examples. He may have seen some evidence, but he's not telling us. So (not for the first time) it's left to Tom Davenport to say something useful. In a short blogpost for HBR, he provides a couple of examples of what can be done when the social and structuring aspects of technology are combined (Want Value from Social? Add Structure).

Note: some of the larger software vendors have a stake in several of these areas, and are trying to integrate different product lines. For example, IBM adds predictive analytics and social networking in Cognos 10 (, 25 Oct 2010). Meanwhile, the niche software providers may be developing interesting partnerships and collaborations - Brian Gentle emails me with a note about the embedding of Jaspersoft within eBuilder, a Swedish provider of an end-to-end B2B suite of Cloud Supply-Chain Management Processes, to produce what they call a Strategic Management Tool.

Places are still available for my Organizational Intelligence Workshop on December 8th.


  1. My article cites the need to tie enterprise 2.0 technologies to specific business processes as a challenge. I didn't pretend to have a solution to the problem. The only example for how to do so that I listed Nielsen's use of innovation management. Nielsen was focused on creating a process, though, not tying the technology to existing processes.

    The link to Ray Wang's article is related to a different challenge: reducing rather than increasing information overload.

  2. Interesting information! provides "Multiply ERP" Software for small to medium level businesses. Software is useful for Hose assembly manufacturers and distributors to manage or plan resources as well as inventory.