In my last post, I introduced the three pillars of the REAL logical architecture of Business unIntelligence. To review briefly, the central, process-oriented data pillar contains traditional operational and core informational data, fed from legally binding transactions (both OLTP and contractual). It is centrally placed because it contains core business information, including traditional operational data and informational EDW and data marts in a single logical component. Machine-generated data and human-sourced information are placed as pillars on either side. The leftmost pillar focuses on real-time and well-structured data, while the one on the right emphasizes the less-structured and, at times, less timely information.
The concept of the pillars emerged from my struggle to clarify the concept of big data and how it relates to the data that businesses have been collecting since we began automating the processes of first running and later managing the business via computers some 50 or more years ago. The pillars are, to a first approximation, delineated my processing and storage concerns. However, there is a more fundamental but less obvious distinction that relates to the sources of the data and information stored and used. These sources are represented by the four clipped boxes at the bottom of the picture.
Let's start in the middle with transactions. Although we often think of transactions in a technical context, more fundamentally they stand for the legally binding agreements (or valid steps towards such agreements) on which all business is based. When we recognize this, we see that traditional operational and informational systems are designed to collect, manage and monitor the contractual and formal information of the business. Before the emergence of big data, we seldom considered in any detail what happened before such transactions occurred.
This thinking leads to the three fundamental information/data sources at the bottom of the picture. Measures and events come from the physical world of machines and show ongoing conditions (e.g. temperature, velocity, location) and changes in conditions (e.g. an acceleration, a button pressed, a call ended). Messages are human-sourced communications in text, voice, image or video that represent something that one person wants to share with another. All of these can and do generate transactions when processed in traditional system. Big data simply records them and analyzes them as "ambient data", as Dale Roberts calls it in Decision Sourcing. Because it precedes transactions and, in many cases, does not lead to any transactions, such information leads to a very different view of the world than our traditional systems provide. Where it precedes transactions, we can do predictive analytics about how the formal business will be affected. When it is completely unrelated to transactions, we can gain new insights into reality that can drive true innovation.
I'll talk more about insight and innovation in my next post...
I will be further exploring the themes and messages of "Business unIntelligence: Insight and Innovation Beyond Analytics and Big Data" over the coming weeks. You can order the book at the link above; it is now available. Also, check out my presentation at the BrightTALK Business Intelligence and Big Data Analytics Summit, recorded Sept. 11 and "Beyond BI is... Business unIntelligence" recorded Sept. 26.
Upcoming conference appearances where I can share more are in Amsterdam at the Business Analytics Congress on Oct. 9 and in Los Angeles at SPARK! on Oct. 15-16.
Posted October 4, 2013 3:03 AM
Permalink | No Comments |