Follow Datanami:
March 13, 2013

IBM: Five Use Cases Driving Big Data Adoption

Isaac Lopez

There are at least five game changing use cases for big data that will drive the adoption and implementation of the trend in organizations throughout the world for the foreseeable future.

In a recent talk, Inhi Cho Suh, VP of strategy at IBM, shared information from customers and implementations that they’ve witnessed over the past two years to provide insight on how big data installations are shaping up in the field.

IBM says that big data is about combining large data, both internal and external together to influence decisions made with your operational data that actually affect outcomes and improve organizational performance.  So with this definition in mind, we examine what IBM believes are among the top compelling uses of big data:

 

Start — Big Data Exploration–>

 

 

#1 Big Data Exploration

IBM references big data exploration as “exploring and mining big data to find what is interesting and relevant to the business for better decision making.  This means that individuals are able to have a federated view of the organization’s data within the context of their position – and explore it at will. 

“Part of this is providing self-service capabilities to everyone in your organization in order to be more data driven,” says Suh.  This type of data access allows employees to create a connected picture, and can be used for everything from new product research, researching customer profiles, pricing optimizations, etc.

The essence of this, suggests Suh, is giving the organization’s people a portal for discovery in order to become more data driven.

Next–Enhanced 360 Degree View of the Customer–>

 

 

#2 Enhanced 360 Degree View of the Customer

Historically, says IBM, the way organizations manage customers has been through a master data management system, a customer warehouse, or a CRM application of some sort. Using big data technology to enhance your view of the customer means connecting all of the picture of the data that exist in the organization about a particular client and then augmenting that with additional data that is actually happening. 

Using life insurance as an example, Suh describes how a customer might be more willing to switch policies during a life changing moment, such as having a new baby, an accident, a graduation, a new job, etc.  Historically, companies would use macro and micro segmentations based on such things as population profiles, where datasets are created using such fields as jobs, salary, age, health, family size, etc.  But with an enhanced 360 degree view of the customer, you’re actually trying to listen to the conversations that are happening in the market in a relevant way that engages the customer at times when they’re most open to that engagement.

Next – Operational Analysis –>

 

 

#3 Operational Analysis

Sensors are everywhere, and becoming increasingly prevalent in our world.  Being able to gather all of an organizations machine generated data and apply analytics to it provides companies with the opportunity to optimize their operations to reduce costs and increase efficiency.

This type of analysis gives an organization two types of capabilities, says Suh.  First, it can provide organizations with the ability to monitor and analyze their systems in real time.  An example might be a GPS sensor on a bus tracking routes and times for analysis.  The second type of operational analysis, say Suh, is pooling large amounts of data and then running deep analysis on it.

The aim with operational analysis is to identify potential anomalies and opportunities as they occur, and increase operational efficiencies, including the avoidance of service degradation or disruption.

Next — Data Warehouse Augmentation–>

 

 

#4 Data Warehouse Augmentation

Think of data warehouse augmentation as “Enterprise Data Warehouse 2.0,” says Suh in describing this aspect.  “What we’re thinking is exploiting technology that advances the value of your existing data warehouse.”

This could be any number of augmentations, such as adding stream capabilities or unstructured data sources.  It could be optimizing the data warehouse storage and providing query-able archives.  It could be adding a cognitive computing solution such as Watson to help with discovery. 

“In this big data world, you have to assume that the questions the business side wants to ask today will not be the questions that they want to ask tomorrow,” says Suh.  Through data warehouse augmentation, an organization can provide the tools for a pre-processing hub for modeling solutions, a queryable archive, as well as exploratory analysis for discovery. 

Next — Security Intelligence Extension–>

 

 

 

#5 Security Intelligence Extension

The last of their five “game changing” drivers for big data use, says IBM, is Security intelligence extension where you’re using the data in motion for crime/fraud prevention.  In this case, says Suh, you’re using the data to find associations, uncover patterns and facts, and maintain the currency of information. 

The directive for this use case is to discover new threats sooner, be able to take action (both preventative and reactionary) in real time. 

Se the entire video here:

 

 

Related Items:

Watson: Coming to a Cloud Near You

SAS Report: Analytics Innovators Trump Mere Adherents

Juneja: HPC, Cloud, and Open Source the Nexus of Big Data Innovation

Datanami