BI City

BI City >>  Expert  >>  

Haranath gnana

"Subtle Hidden Trends in Business Intelligence and Analytics "

Haranath gnana
Practice Area Leader
Saama Technologies
Haranath is a Practice Area Leader at Saama Technologies whose whole career spans over 18 years in information systems, with the first five years in Industrial Process Automation at conglomerate ABB. His technology skill sets include the development of complex web applications, portals and content management systems, along with specialization in information management and business intelligence (BI). His expertise includes a broad swath of the information management lifecycle, from formulating BI roadmaps and strategy to building operational realities, in the process enabling enterprises make informed decisions. At Saama, he was key in creating the thriving Life Sciences practice, which he now oversees, along with the Health Care practice.

At Saama, he has helped drive "labs" initiatives that explore cutting-edge technologies, primarily in open source, to derive additional value for customers. He built Savii (Smart Audience Viewing Intelligence & Insights, an enhanced variation Nielsen ratings) and PSL (Predictive Store Locator, which small retailers use to identify optimal store locations) by leveraging syndicated public data sets, along with open source data mining and predictive modules. He received a patent for his pioneering work in PSL. He also helps clients adopt and leverage cutting-edge technologies such as BI appliances and mobile BI. Haranath has been published in leading outlets such as,, and Office of the CIO.

He completed his MBA with Honors from Lucas Graduate School of Business at San Jose State University, earned his B.S. in Electronics Engineering from Bangalore University. He is a Certified Business Intelligence Professional (CBIP) in Leadership & Strategy at the Mastery Level from TDWI.

Business intelligence (BI) trend articles from a variety of authors are published fairly frequently. For this reason, I initially resisted writing another but noticed that not many folks were talking about or even mentioning anything beyond the current usual suspects of big data, the cloud, unstructured data, Agile BI (the iterative approach), self-service enablement, etc. While these do represent big disruptions in the BI landscape today, I want to share some insights about a couple of trends that I believe can have an equal or even bigger impact on how business intelligence is consumed.

Actionable versus Action-Enabled Business Intelligence

Although “information-enabled decisions” is a philosophy that has been discussed and promoted widely in the recent times, information-enabled decisions have not yet become the mainstream SOP (standard operating procedure). Now that many business leaders are mobile and/or remote, information-enabled decisions have become an even bigger challenge.

With the entrance of iPhone a few years ago, it appeared that the mobile workforce would have access to entities beyond email and calendar that the BlackBerry had made so popular. The iPhone interface allowed for information and insights to be presented to the field force, better enabling the field force, and these devices gained pretty quick broad adoption. But the form factor was still too small for effective use of information. With the entrance of the iPad, that changed completely. Now there is a platform that allows the mobile workforce to gain access to insights without a constrained screen size. Given that enterprises are spending a lot of effort generating and presenting relevant information nuggets / insights, there is a clear push for “actionable intelligence.” Now available to the field force on their tablets, this becomes “actionable intelligence OnTheGo.”

However, the added twist to this is what I term “action-enabled business intelligence” or, adding the mobility aspect to this, mobile action-enabled BI (MAEBI). (I pronounce it mayeebuy). Here users are presented with not just the “actionable insights,” but also the relevant transactional screens that enable them to make a decision and take an action.

For example, think of a marketing executive running a campaign. With mobile-enabled business intelligence, he / she would have access to all of the details of how the campaign is progressing. Of course, this assumes that the plumbing between the campaign management system and the BI repository has been set up, which has become quite common in enterprises today. However,  when the presentation of the information to this marketing business user is not just the BI insight but also presents the BI information on the same screen in the context of the actual possible actions this person could take, then the person’s decision-making process – including the action taken – becomes a lot more efficient. This is the concept of “action enabled business intelligence.” This is also sometimes referred to as “BI Composite Application Screens.” Now the iPad makes this possible for the business users on-the-go – making this the “mobile action enabled BI.”

Figure 1 is a mockup of a screen of such an application on an iPad. The top section presents Actionable BI Insights while the bottom part is content from a CRM application. Given the current maturity of this platform, this application can be designed intelligently to present the right CRM application modules when looking at certain BI components and vice versa.

Figure 1: BI and Transactional Application Components on the Same Screen

All of the top tier BI vendors are already providing a platform / framework to support this (i.e., combining BI insights inline with transactional system forms for actions). Although in its infancy today, this will allow our highly mobile workforce to make decisions based on relevant insights and take actions all on one screen, on the go. This opens an entire new world of possibilities, adding a whole new perspective to how the mobile workforce could operate.

Predictive Analytics for the Masses

This seems like the next wave beyond the “BI for masses.” Predictive analytics / data mining as a technology has been available to mainstream commercial organizations for several years now, but it has still been the tool of a small team of data specialists (statisticians, etc.) tucked away in some corner of the enterprise. These folks have the job of predicting the future by analyzing huge amounts of data using compute brute force. These teams use very sophisticated software and mathematical algorithms and try many different models and data massaging techniques to come up with a final set of models.

AlI of this, I believe, remains unchanged for now, but what I believe is changing is how these models are being made available to the front line business teams to look at different scenarios in a self-service capacity. Many of the tools today are starting to provide a way to integrate predictive models and present these as additional capabilities for analysts to utilize in their discovery processes. Predictive modeling has gained a lot of popularity in the marketplace over the last several years, but making these models available to a much larger user base (i.e., the front line teams) is going to significantly accelerate the further adoption and use of this technology.

Figure 2: Segmentation Process with User Selectable Parameters

Figure 2 shows how an analyst could leverage a segmentation statistical model almost like a custom function within the BI application. This allows for the users to be able to play out scenarios before finalizing their segmentation buckets. The key is to ensure that the technical complexity in building the statistical models, testing and validating them is still managed by a core team of experts, and enabling these validated models to be available as custom functions within the BI environment. This allows the core team to focus on building and validating better models while offloading the tasks of iteratively tuning/refining the models for specific needs of the BI end users.

As with any technology its importance and significance is directly tied to adoption. This change in allowing for these models to be leveraged within BI applications significantly increases the adoption of these extremely powerful technologies, further strengthening their significance and setting a stage for its rapid growth.

In today’s ultra competitive world with the ever-increasing budget pressures, we will continue to see rapid innovation in improving both the efficiency and effectiveness of the decision-making process. The two trends presented in this article are evidence of the growing demand for these capabilities.