Mission and make up of a data sciences group

The market is abuzz with data trends like big data and data science, and there are more to come. Numerous organizations across the world are trying to find the best possible way to establish an effective data group. However, organizations face various challenges in setting up a best-in-class data group, and those who have somehow managed to set one up face issues sustaining it. Hence, it has become very critical to analyze and understand the factors responsible for making this task challenging.

One of the key reasons for the nascent death of such a data group is the incapability of data organizations to continuously showcase their real potential to a business. These groups usually have technologists and evangelists who build on existing success and take on more volume and velocity of data, variety of data types—including structured, unstructured, and semi-structured—and provide real-time data processing and analytics capabilities.

The organization keeps the group and the hype stays on for a while until interest dwindles. Due to lack of adoption by business to sustain growth and interest, such data groups slowly cease to exist.

This article highlights the ways that help with the setup and sustenance of a data organization—focusing on five must have areas for this group. Going with the current trends, this group is referred as the data sciences group.

The Mission
The mission of a data sciences group spans a variety of priorities—providing protection, value, predictability, accuracy, and easy access.

The first goal is to provide solutions that help eliminate the vulnerability of business, such as loss of customers, inappropriate transfer of sensitive data, and market threats.

Secondarily, they should offer valuable solutions to keep a constant watch on the multi-channel customer voice and adapt to the pulse of a business’ biggest asset, the customer.

Thirdly, it should provide solutions to help businesses make faster and more accurate decisions using prediction models to eliminate human error and create a data-driven organization with veracity.

It should also offer solutions to help lower a business’ total cost of operations es using disruptive technologies and eliminating technical debt.

Finally, it should provide solutions that are be easily accessible and available anywhere, everywhere, and any time.

Group Composition
A data sciences group should be made up of a business analyst; a data analyst; technical resources with knowledge of mobility, visualization/user experience, and big data technology; and a business sponsor.

Every project this group executes must have a sponsor from the business who spends time ensuring that the solution being developed delivers value.

In terms of deliverables, the business analyst must be able to create a product requirements document for the solution, road map, ROI, and benefits to the business. The data analyst must be able to point to source and the target data deliverables and be responsible for the data quality of the solution. The technical resources own the technical solution and architecture. The business sponsor takes responsibility the deliverable to make the solution operational in the organization. In terms of training, the data sciences group needs to constantly learn and keep pace with the technological advances so that the solutions developed are innovative.

Data Sciences Group’s Evangelization
A data sciences group can only be evangelized by the ultimate operational and analytics users of the solution in question. They are the people who can keep this group from disappearing into oblivion and they do it by integrating the newly developed solutions and adopting them. The most important thing to note is that this group should not be confused with the enterprise data warehouse group.

The article was originally published on Software Magazine on November 26, 2014 and is re-posted here by permission. 

Kumar Ramamurthy

Vice President, Chief Technologist - Enterprise Information Management (EIM) Practice, Virtusa. Kumar has over fifteen years of experience in enterprise data architecture, database related technologies, software platforms and architecture assessments. Kumar is primarily involved in consulting engagements and assessments for Virtusa at existing and new EIM clients. He also has overall responsibility for delivery assurance from the EIM practice at Virtusa. Kumar has proven ability in consulting, selling, driving, delivering large scale EIM development/maintenance projects both in the enterprise and ISV spaces, while ably bridging the technical and business worlds to ensure the delivery of the best, most accurate business solutions to clients. He is particularly knowledgeable in Kimball and Inmon related DW architectures. He is adept at Data Integration, BI, Data Governance, Database Performance tuning, data modeling and MDM areas. Kumar has a Masters in Computer Science from Bharathiar University, Coimbatore, India. When he is not creating EIM solutions, Kumar spends time with his son and enjoys golf at his Arkansas home.

More Posts