Big Data at Work – Where to Begin?

Analyzing data to improve organizational outcomes is not a new concept. From “Decision Support” to “Business Intelligence” to “Big Data”, the idea of data analysis is not new. What is new is the sheer volume, types, and increased availability of data to improve organizational outcomes.

With the number of active wireless connected devices currently exceeding 16 billion, Big Data deserves management attention. In his recent book, big data @ work, Thomas Davenport explores the opportunities available from Big Data and explores considerations for developing a strategy for implementing Big Data in an organization.

Determine Big Data Objectives

The first step is to determine organizational objectives for Big Data initiatives. These objectives will ultimately drive the planning, process and management of Big Data projects. Possible objectives to consider include cost reductions, process improvements, identification of new product offerings and improvements in internal decision processes.

Cost Reductions – Big Data technologies such as utilizing Hadoop clusters instead of traditional databases can save thousands of dollars annually. Of course, new technologies need to be evaluated on issues important to your organization, such as reliability, security and ease of management.

Streamlining processes can also result in time and cost savings. Understanding the potential business performance improvements requires that your data scientists or IT personnel work closely with the area managers for the process involved.

Decision Improvements – Adding new data to existing decision models can have significant impacts on supply chain management, risk management, pricing decisions and customer behaviors. Organizations are adding data points to help evaluate internal questions such as:

“What offers should you present to a customer? Which customers are most likely to stop being customers soon? How much inventory should we hold in the warehouse? How should we price our products?”

Additional data points may be gained by using new tools to allow information gathering, such as converting call center recordings into structured data or capturing and analyzing customer “journeys.” Analyzing customer “journeys” involves tracking a customer or potential customers’ website clicks, page views, transactions and customer support interactions. Each set of actions is correlated to problems or purchases. This analysis can provide information for decision support which was never available before.

Improvements in Products and Services - New product and service offerings are quickly becoming an organizational goal of Big Data. Online product offerings based on data analysis may be easier to see from companies such as LinkedIn and Google, but offline businesses also have countless opportunities.

Determining Responsibilities

Once the key objectives for Big Data have been determined, it is necessary to outline responsibilities for who will oversee the project. There are 2 phases to consider: “discovery” and “production.” The discovery phase of each initiative involves determining what data points are available in your data and how they may support achievement of each objective. Production involves using the information obtained during discovery and implementing it into the organizational processes.

If cost reduction is the main objective then decisions regarding adoptions of software, visual analytics and other technologies are primarily based on technical and economic factors. The discovery phase for a cost reduction project is generally led by IT innovation personnel and the production phase is implemented by IT architecture and operations.

The discovery phase of decision improvement initiatives is directed by the relevant business unit or functional group. Implementation is then led by the executive management of the functional group.

Product and service innovation discovery phases are managed by R&D or product development. Production phases of new product/services are led by product development or product management.

The opportunities available from Big Data for any type of organization can be limitless. Knowing how to start can lead to an enterprise strategy for implementing Big Data and achieving organizational goals.

For more ideas on implementing Big Data, also reference our previously published article, Best Practices for Big Data Adoption.

facebooktwittergoogle_plusredditpinterestlinkedinmailfacebooktwittergoogle_plusredditpinterestlinkedinmailby feather