BI System Design incorporating Wiki and other Web 2.0 Components

Suddenly collaboration is flavor of the month, or year anyway. Customers are re-designing products, buyers are guiding the choice of other buyers, repair and service people are specifying work procedures, those with spare time collaborate in wikis-everything; and maybe the lunatics are running the asylum? But what of Business Intelligence systems, what is their place in all this?

I’ve been preaching the utility of collaboration an essential element of BI for 20 years now. Maybe it’s going to happen at last?

But, even with the current enthusiasm, it won’t happen at Youtube speed. Collaborative BI is more complex than just loading some videos or other data into a category for others to retrieve. And, supposedly, corporate people don’t have time to surf the Intranet, let alone the Web, looking for relevant stuff.

Therefore, it will take time; and remember the old timer’s adage: “You can tell the pioneers; they’re the ones with the arrows in their backs!”

Here’s a set of axioms that I believe are relevant if we are to succeed in this collaborative endeavor, all of which raise barriers, some large, some not, it depends on the business environment:

• Corporate people who come across Web 2.0 style intelligence often don’t know its value, and whom to tell
• They usually only have part of the story anyway
• They often lack the background to be able to assess implications
• Supplying intelligence to Web 2.0 style repositories or applications is time consuming, and may not be at all rewarding to the author, only to others
• Intelligence can’t be searched for, or be subject to push messaging and alerting, if it’s not categorized
• Categorization must conform to a corporate standard vocabulary, or it will not facilitate sharing and collaboration
• All BI items indexed by a category are not of equal significance or value; some may be critical to the business, others routine news that’s already well known
• The high value items ought be separated from the dross and given wider audience, or personalization will be ineffective; but how to do this?
• BI is useless unless the recipient can assess its implications, and often this requires additional input of BI, or experience, from other people or sources – the collaboration imperative
• Corporate collaboration raises infinitely more cultural and behavioral red-flags than Web 2.0 practitioners could dream of; see earlier post.

Nonetheless, I’m sure that we’ll see increasingly effective means for accommodating the issues raised by the above points. I have suggested some design principles in another earlier post, but the rapid evolution of wiki style knowledge creation, with the attendant blog explosion, is opening up new opportunities.

I believe that the issue of bringing new knowledge to the attention of the right people, personalized distribution as the knowledge is created, will remain a substantial BI issue. My ideas on this will be the subject of a later post.