Insights and opinions on Information Technology from George Tomko, a veteran CIO turned renegade consultant.

Rant:

Does Business Intelligence Require Intelligent Business?

By George M. Tomko

As I started to develop this blog post, it occurred to me that my working title, “Does Business Intelligence require Intelligent Business?” might have been previously used in some other publication. So, I Googled it.

majestic-tree-small

I did find some very close variations, but not exactly in the form of the question that the title poses.

I was led to a white paper, “business intelligence is intelligent business”, by Gerry Davis, Regional Managing Partner, Asia-Pacific for Heidrick & Struggles. In the opening paragraph, the problem is summarized thusly:

“Collecting information about customers is relatively easy. Analyzing customer information for potential cross-sells, increased revenue streams, and improved service is more challenging. But getting the information to the front line in a timely manner and thus providing further competitive edge is proving increasingly difficult for many corporations.”

As we look at this statement, there are three main points: 1) collecting is “easy”; 2) analyzing is hard; and 3) disseminating it is very hard. Perhaps a bit oversimplified. But, in reality, most users will need this to be oversimplified to be able to overcome all their biases about IT, systems in general, any extra “work” that will automatically be assumed and fears about job security. This is said this way, not to be unkind, or even to be negative, but to make sure that the focus is on the right “problem”.

Like the human brain, which, in a lifetime, is barely tapped for 10% of its ultimate potential, the organization “brain” is woefully underutilized. A number of studies and surveys have consistently shown that enterprise resource planning (“ERP”) software, such as SAP, Oracle, etc. are underutilized.

As Albert Einstein was once quoted, “Computers are incredibly fast, accurate and stupid. Humans are incredibly slow, inaccurate and brilliant. Together, they are powerful beyond imagination.” Such a proclamation will only be true if there is a significant change in the approach that knowledge workers take to their jobs.

Mr. Davis suggests that organizational change may be the answer:

“As business intelligence divisions spring up in organizations across the globe, the same question is being asked: “How can we unlock the value of our data?” At Heidrick & Struggles, we believe that the answer lies in implementing an appropriate organization structure and in identifying and appointing the right executive — someone with superb business acumen combined with a sound technical understanding — and tasking them with delivering real business intelligence.”

In a very weird way, I actually agree and disagree, simultaneously, with this statement. For a company like H&S, an organizational solution is going to feel like the absolutely right thing to do. After all, that is one of the main things that they do. For organizations that commit to this approach and throw the full weight of their senior management commitment (and funding) at sustainable levels, it may work out very well.

But, there are quite a few brilliant people who are many times smarter on the subject of business analytics and business intelligence that will tell you that knowledge is a process that begins with data (“D”) and moves along a progressive transformation process into information (“I”) then knowledge (“K”) and ultimately, wisdom (“W”).

If the CIO owns the “D” and the “I”, and the CKO owns the “K”, who “owns” the “W”? Is there a collective ownership of the wisdom of the organization? Will we be seeing “CWO”s in our future?

It was back in 2000 that, with great assistance from Bill Odom, I drew the following picture. At the time, I was a first year CIO trying to stand up a new joint venture. With as close to a clean sheet of paper as one could get, we were architecting the mission of IT from end-to-end. And we got the chance to build it.

dikw-chart-small

Since then, a lot has changed in this world. But, some things stay the same. No matter what the advances are that enable us to do things better, faster, cheaper, it still comes down to people and how people do things. Remember Einstein’s quote above.

Another brilliant thinker that captures the essence so well is Peter J. Thomas who writes in his recent blog post:

Business Intelligence is not just about technology and cannot be effective in isolation. To live and breath it needs to be part of a broader framework covering the questions that its users need to answer, the actions that they take based on these answers and the iterative learning that occurs in the process.”

At the end of the day, there is nothing I would like to see more than knowledge-based business processes executed by knowledge-savvy workers (and anyone else in the value chain) benefitting knowledge-enabled customers.

But, to see that idea realized, we have to understand and deal with the three main challenges that we previously identified.

1) Collecting information about customers is relatively easy. CIOs have been pretty effective at collecting data through transactional and automated data collection. So good, in fact, that most organizations are “drowning“ in data. In most cases, it is in multiple formats, fragments, repositories etc. Master data must also be dealt with, but, given the unbelievable chore of maintaining it, master data is often an “attic” filled with junk mixed in with the useful stuff.

2) Analyzing is hard. The issues cited in item 1 give the first clue why this is so. First, as previously noted, we are drowning in data. Second, the data that describes the data, i.e. “Master Data” and gives it purpose and context, is often an overburdened mess. Efforts at standardization of form and meaning collapse under their own “weight”. It certainly does not help that many companies are actually companies of companies. There are so many pieces that are pulled together via acquisitions, “carve-outs”, reorganizations, etc. that base lining and normalizing company data so that historical analysis can be done is like boiling the ocean. At a minimum, this becomes an activity that I call “data mashing”.

3) Disseminating it is very hard. While products are evolving to provide robust tools that will essentially offer “composite” applications that sit on layer(s) above the legacy applications and data stores, few organizations and their business processes are set-up to meaningfully “plug-in”. What will happen to all of the spreadsheets!?

Having said that, I am not sure that the answer lies with a new executive in the organization, driving the organization to come up with its business intelligence. Indeed, it is the remnants of the middle manager and senior analyst layers that are closest to the actual in-place and functioning business processes and work-flows. They know the business rules. They know how things “really” work. They know how to handle situations that come up that require judgment and understanding of the facts of each case.

Too many businesses play with fire by running in the margins with their knowledge workers. Thus, a problem exists where these folks are highly marginalized and “one-deep” on the resource chart. They are essential members of the business teams for everything that the business wants to do. What happens if they get hit by the proverbial bus, retire or quit?

Looking through this lens, creating a new silo, or a new organization matrix that adds a new “boss” for these individuals does not, in my opinion, get the organization the knowledge and wisdom that it needs to sustain and grow.

Ironically, this is one of the major business problems that service-oriented architecture (“SOA”) and business process management (“BPM”) is/was supposed to cure. The problem with SOA and BPM is that they are BIG ideas. BIG ideas usually require BIG money, time and resources. Of course, the expectation is also for “BIG” results, which would make it all worthwhile.

Going beyond this point would far eclipse the scope and intent of this particular “Rant”. Getting back to the question that was posed in the title of this post: “Does Business Intelligence require Intelligent Business?”.

Organizations are structured, for the most part, in verticals a.k.a. “silos”. Having efficient, well-designed, ubiquitous cross-functional processes that work is still a rare find. In that regard, a new executive and organization that creates the silo-to-silo (i.e. “horizontal”) channels and portals may have promise. But, hierarchy still rules the day and there is personal “safety” in these turbulent and vulnerable times for incremental approaches.

This sort of “incrementalism” does not create a noticeable increase on overall organizational business intelligence. To the extent that transformational change brings the IQ of the organization to new heights, then there is hope for creating an intelligent business.

Is there a chicken or the egg story here? What comes first – the chicken or the egg?

In organizations, what comes first – business intelligence or intelligent business?

There is, of course, no absolute answer. It could be said that if not an intelligent business then what worth business intelligence?

What do you think? Please offer your comments.

©2009 George M. Tomko All Rights Reserved