In the last part, I discussed the interplay of data and usage of data and how it creates the business knowledge. Business value of Big Data does not come simply from deployment of an initiative. Business value is realized from the usage of this data. In this final part of the series; I will discuss the dynamics of the interaction of data and business questions; creating insights addressing an existing pain point or leading business to frame a more focused follow up question.
Data made available by a big data initiative can be categorized as data that companies know they have and data that companies did not even realize they had we can call this unknown data. Example of first category is customer data collected from call center activity or customer interaction on company website. Example of unknown data may be in the form of leverage known data to create a new trend analysis or customer stratification analytics. This use of data is not evident when the available known data is viewed at more granular level, similar to focusing on individual tree not realizing that tree actually belong to a forest. In order for business people to take advantage of both categories of data, they must frame appropriate questions from the data.
Questions framed by business could be clearly defined, for example what is the quarterly customer satisfaction rating. Well defined questions are quiet easily answered from the known data as these questions typically target data people know is available. So business people knew what data was available and what question it could support, they ask these defined questions to drive knowledge by calculating a customer satisfaction score without too much difficulty, as an example.
Business could also ask a question from the data where the data itself is unknown for example business may ask effectiveness of a proposed email marketing campaign. Here what business has asked is well defined but however data about a future campaign is not known. However; IT can provide this data by manipulating the existing known data. This is the iterative nature of a business intelligence initiative that I talked about earlier. What IT would probably do in this case is first they will try to find the meaning within the available data by creating new analytics, trends and aggregation. In case there is not enough data available to drive the analysis IT might look at other source data and may bring in new types of data to answer the question. In this case what IT has done is taken unknown data that exists somewhere in the organization and manipulated this data so that it became known to the business; allowing business people to ask questions of this data. This shows that IT can find data to answer well defined business questions as long as question itself is clear cut. It is still up to business people to ask the right probing questions to create new knowledge.
Bigger challenge for Business Intelligence in general and Big data in particular is dealing with the situation where business does not even know what question to ask irrespective of if data is known to them or not. This is best understood using an example. Let us take the example of Automatic meter reading, or AMR technology. Utility meters equipped with this technology collect consumption, and diagnostic data. This technology may have been deployed by the providers to save on the costs and billing delays associated with the manual meter reading. These devices create a deluge of data that can only be analyzed using Big Data and MapReduce technology. IT may make all this data available as a Big Data initiative.
At this point business people may not realize that this data can support a whole set of new questions; not thought possible from simple manual meter readings. To drive value out of this real time data, IT and business will have to go through a cycle of iterations. Business may start by creating a hypothesis about this data; IT finds supporting data to the hypothesis. Upon receiving this analysis business will ask secondary order questions. IT finds the answer to these questions so on and so forth. This process has to go on till a point is reached where new knowledge is created; for example this data helps predict energy consumption patterns allowing energy traders to make purchases of future contracts based on this new knowledge.
I like to conclude this discussion by reiterating the allure to business that IT in itself can provide solutions to address business problems. Assumption is that business leaders will be able to make better decisions by having access to more data of better quality. As I have clearly shown above that technology deployment only provides a capability to the business. The quality of business decision depends on how this data is leveraged by the business leaders. Better decisions and business insights are result of a combination of factors. Enough, timely, and quality data is a good starting point. Other more critical factors are things like corporate culture and the capabilities of the business executives.
Companies will have to find ways to leverage the huge amounts of internal and external data to their advantage. Here risk may be that organizations adopt traditional approaches; instead of adjusting their collective mind-set leading to a possible paradigm shift when approaching Business Intelligence and investment in Big data technologies. Find book on subject here.
This entry passed through the Full-Text RSS service — if this is your content and you're reading it on someone else's site, please read the FAQ at http://ift.tt/jcXqJW.