Cognizant 20-20 Insights

Mastering Big Data: The Next Big Leap for Master Data Management
By leveraging big data, organizations can provide a true 360-degree view of critical master data entities; meanwhile, master data can help convert information gleaned from big data sources into actionable insight.
Master data management (MDM) is critical to the success of many big data analytics projects. Why? Because it encompasses people, processes and technologies that ensure the accuracy, uniformity and uniqueness of data entities shared by multiple business units within an organization. (See our white paper Innovations in MDM Implementation: Success via a Boxed Approach for additional insights.) Big data, on the other hand, refers to the large quantities of humanor machine-generated bits and bytes that are generated at high speeds from multiple sources, including not only posts and comments on social media but also images, videos, PDFs, RFID data, point of sale data, call center comments, data generated by enterprise systems and more. A big data analytics strategy involves collecting this data in order to create meaningful information. All this data, in various shapes, sizes and forms, represents valuable information that needs to be standardized and stored in a meaningful way within a central repository accessible by all business units. MDM provides a trusted, central point of storage and access to this information. This whitepaper describes how master data management and big data analysis need to work hand-in-hand to provide a 360-degree view of an organizations critical data entities. (While this paper is largely focused on social data and customer MDM, there are many other applications of big data not explored here.)

The Relationship Between Big Data and MDM

MDM provides a meaningful context to big data and a panoramic view of critical data entities, such as product, customer, employee, agent, location, etc. (see Figure 1, next page). With all this information stored in a central trusted repository, all business units have access to a single version of the truth, on demand. In the absence of such a master data hub, records are typically stored in silos, such as loyalty systems, CRM systems, POS systems, Web portals and big data repositories. Such systems usually have their own version of the truth, causing headaches that ripple across the organization, from substandard customer service and bad business intelligence, through impaired operational efficiency (see Figure 2, next page). While MDM is crucial to the success of many big data projects, it is important to note that not all big data initiatives require an MDM solution. MDM is needed only for big data initiatives that involve

cognizant 20-20 insights | may 2013

A 360-Degree View of Jane Doe

Unstructured Internal Data
Call center conversations Point of sale data

Third-party Information

Nominal Information

Credit score Professional affiliations Memberships

Name: Jane Doe Address: 1 Acme Road, Kiwi NJ 00000 Phone: 555-555-5555 Age: 23

Conversations Reviews Blogs Images Videos

Unstructured External Data

Figure 1

master data entities, such as customer, employee, agent, product, location, etc.

The Scope of Big Data and MDM

When data is extracted from most big data sources, it contains two parts. The first part is entity attribute-level data, or data that can be used to identify an entity. The other is transaction-level data that is generated by each entity. For instance, if the big data source is Facebook, the two types of data include:

The big data extraction layer will pass the attribute-level data to the MDM system and the transaction-level data to the big data hub, to be analyzed by a big data analysis tool. After analysis, the brand advocate score, purchase intent, etc. will be passed to the MDM system to create the 360-degree view of the customer. To help define the scope of big data, consider the following:

MDM will act as a repository of the following >> Attribute-level data. >> Information derived
data.

Attribute-level data: Facebook ID, phone

number, name, address, etc. comments, likes, etc.

information types obtained from big data sources:

Transaction-level data: Conversations,

With or Without MDM

from transaction-level

CRM System CRM System

Basic customer information, campaign information, targeted mailer, couponing information. Basic customer information, campaign information, targeted mailer, couponing information.

Social Data Big Data

Social conversations, whether the customer is a net promoter, on the verge of switching, network reach, etc.

POS System
Customer store purchase information and the information entered in loyalty card.

Social Data Big Data

Social conversations, whether the customer is a net promoter, on the verge of switching, network reach, etc.

POS System
Customer store purchase information and the information entered in loyalty card.

No reliable, 360-degree view

Central reliable repository

Web Portal
Web purchase history, information entered during checkout.
?

IVR Portal
Call center conversation, number of problems resolved, dissatised customers.

Web Portal
Web purchase history, information entered during checkout.

IVR Portal
Call center conversation, number of problems resolved, dissatised customers.

Figure 2

cognizant 20-20 insights

MDM and Social Architecture

Permissions /data extraction layer

Facebook

Entity information such as Facebook IDs, Twitter IDs and other extractable entity information is syndicated to MDM.

MDM-on-Cloud

Data cleansing Data standardization De-duplication Workflows Hierarchy, etc.

Twitter

Analytics Tools
Post-analysis attributes like net-promoter ag, at-risk ag, Inuencer ag etc., are passed on to MDM.

+
Company Web site Others

CRM

Social conversations are passed on to the analytics tool.

Social Analytics Tool

Sentiment analysis Lead generation Customer acquisition Net promoter identification Attensity, Pentaho, Googles Social Interaction Analytics, Cognizant Social Prism and many more.

Figure 3

MDM will cleanse, standardize and de-duplicate

this entity-level data and maintain audit trails and hierarchies. will not store entity-level transactional data, such as conversations, transactions, etc.

MDM for Big Data Analytics

To handle the demands of big data, next-generation MDM systems must successfully handle the volume, speed and formats across all data types. Required fundamental changes to MDM include:

MDM

MDM will not analyze big data.

In the absence of an MDM system, both attributelevel and transaction-level data can be stored in big data hubs. But such data stores are meant for heterogeneous, unstructured or semi-structured data that is intended for analytics purposes; hence, they are not optimal for managing shared enterprise master data entities. Additionally:

data model to store big data-specific attributes, such as social media IDs, opt-ins, brand advocate scores, social network hierarchies, product preferences, purchase intent, churn possibility, etc. As previously mentioned, some of these attributes will be fed into the MDM system from big data analytics systems; others will be fed directly from the source. handle the speed at which this data is both added and updated. match and merge processes, especially for social analytics, which is among the most widely applied use cases in big data. The biggest challenge with social analytics is successfully identifying the social profiles of an organizations existing customers. This is especially true since multiple results can be returned when searching for common names. Also, many people use aliases rather than real names in their social profiles, making it even more difficult to identify these profiles. data governance policies involve rules around not only data usage, ownership, etc. but also around big data-specific data governance and data stewardship. MDM

Data models in big data hubs are optimized for analytics, while MDM data models are optimized for entity resolution. A big data hub stores all information gathered
from a big data source; however, these records are neither unique nor cleansed and rectified. Hence, finding an authoritative golden copy can be difficult and time-consuming. will not allow a quick search for entity-level attributes for customer service or decisionmaking purposes. big data analysis. Hence, strong entitlement and hierarchy management are not included.

Improved MDM integration architectures to Advanced

The sheer volume of data in the big data hub

Big data hubs are built for a specific purpose

Advanced

cognizant 20-20 insights

needs to enforce policies and procedures that ensure:

>> A downstream couponing system can readily

>> External data does not affect the integrity of

the internal data. is mined.

>> Data is relevant for the purpose for which it >> Escalation paths exist for data conflicts. >> Privacy policies are enforced.
When Big Data and MDM Combine to Achieve a Business Result
Airlines With enormous customer acquisition costs, customer centricity is the key to profitability in this industry, which makes airline carriers ideal candidates for big data programs.

obtain this data from the MDM hub to identify heavy users of certain products and based on the preferences, opt-ins, etc. of these customers send them targeted coupons.

Fashion Retailers With rapidly shifting styles, fashion retailers need to update their product assortment continuously. Garments that are in vogue one season can go out of style the next, which means retailers must clear their racks using discount pricing strategies. The longer that old merchandise sits in the warehouse, the more it loses value. So its imperative for retailers to sell their merchandise as quickly as possible.

Problem

Problem statement: An airline wants to send

targeted promotions to brand advocates and influencers to increase coupon redemption and positive word of mouth.

statement: A retailer wants to build a complete profile of all customers so it can recommend ideal products and discount offers, at the right time and place, to improve customer service and profitability.

Solution: >> Route a social media campaign through the

airlines loyalty portal to capture customers social media presence. and influencers among these customers and tag them appropriately in the MDM hub.

Solution: >> Implement

>> Analyze big data to identify brand advocates >> A downstream campaign management sys-

big data analysis software to analyze point of sale data, purchase history and data from social media (Polyvore, Pinterest, Facebook, etc.) and create a complete customer profile in the customer hub. data in the customer hub in real time to recommend products, coupons, etc. at the time of checkout or as part of marketing campaigns that send targeted coupons and promotions to these customers.

>> Analyze

tem can readily obtain this data from the MDM hub to identify brand advocates among an organizations customers and based on the brand advocates preferences, opt-ins, etc. send targeted promotions to them.

Satellite TV and Cable Providers These providers can analyze customer channelsurfing behavior for many purposes.

Multibrand Retailers Due to cost pressures, severe competition and perishability of certain products, couponing is among the most commonly used marketing tactics in retail. Targeted couponing enabled by big data analytics would help retailers improve coupon redemption rates and increase profitability.

Problem

statement: A satellite TV/cable provider wants to show targeted ads to customers to increase advertiser ROI.

Solution: >> Analyze

Problem

statement: A retailer wants to identify heavy users of certain products and send them targeted coupons to enable quick inventory clearance.

individual customer channel-surfing behavior to understand which customers are likely to watch which types of ads during which shows. on this intelligence.

>> Enrich customer data in the MDM hub based >> Use downstream ad placement software to
obtain channel-surfing information, along with viewer household information, opt-ins, etc. from the MDM hub on-demand, and based on that insight, target ads to viewers.

Solution: >> Implement

big data analysis software to analyze point of sale data for loyalty card holders, and tag frequent purchasers of a product in the MDM system as heavy users.

cognizant 20-20 insights

Multiproduct Manufacturing Companies Such companies must manage a large number of suppliers and have a 360-degree view of them via a central repository. This is important not only for supplier selection purposes but also for regulatory compliance.

Critical Move-Forward Considerations

While big data is already the next big thing in business-technology, and holds the key to achieving the next level of customer centricity, organizations that want to make sense of huge, proliferating pools of structured, unstructured and semistructured data need to keep the following in mind:

Problem statement: A multiproduct manufac-

turing company seeks a 360-degree view of its global supplier base for quick and efficient supplier selection.

A big data strategy should be a companywide

Solution: >> Analyze

initiative, with complete buy-in from senior leaders. all enterprise systems in any size company (i.e., business intelligence systems, data warehouses, ERP, CRM, campaign management, etc.). Therefore, a proper IT integration strategy is crucial.

available information about each supplier via the media, the Web and from notes transcribed by procurement managers during transactions. to build a 360-degree view of each supplier, including violations cited in the news, delay in past shipments, etc.

Big data will impact not only MDM but almost

>> Enrich supplier data based on this analysis

It is important to establish a clear understand-

Automobile Manufacturers Car makers tend to spend millions of dollars each year on market research for product line extensions or feature enhancements. Big data can be analyzed to understand what customers, partners and competitors think about a car and more.

ing of the expectations from big data analytics. The social Web holds terabytes of data for most brands; acquiring, storing and attempting to mine all of this data is not only impossible but also unnecessary. brands need to maintain a balance between targeted promotions that consumers find useful vs. targeted promotions that consumers may find intrusive or an encroachment on their privacy.

Finally,

Problem statement: In an attempt to expand

its business, a car manufacturer seeks to aggregate all feature-related commentary about its models and use these insights to shape future vehicles.

Final Thoughts
In this era of customer centricity, superior customer service and operational efficiency are two major differentiators that can set a company apart from its competitors. Big data analytics can help improve customer service, develop operational efficiencies and increase profitability. However, big data analytics cannot provide actionable insights unless the data and related information derived from big data sources can be traced back to the appropriate entity. Therefore, to get the full value from big data initiatives, MDM is not just an option but a critical necessity.

Solution: >> Analyze social media commentary, calls to

the call center related to each model and data from automobile monitoring systems, if available. uct information management system, including features to be added in the next model, so that all information about the vehicle resides in the same place.

>> Enrich the data about each car in the prod-

About the Author

Reshmi Nath is a Business Consultant within Cognizants Customer Solutions Practice, focused on master data management. She has seven-plus years of experience in the technology and consulting industries in the U.S. and India. Reshmi earned her M.B.A. in marketing and finance at Michigan State University and a bachelors degree in computer science at Bemidji State University. She can be reached at Reshmi.Nath@cognizant.com.

cognizant 20-20 insights

About Cognizant
Cognizant (NASDAQ: CTSH) is a leading provider of information technology, consulting, and business process outsourcing services, dedicated to helping the worlds leading companies build stronger businesses. Headquartered in Teaneck, New Jersey (U.S.), Cognizant combines a passion for client satisfaction, technology innovation, deep industry and business process expertise, and a global, collaborative workforce that embodies the future of work. With over 50 delivery centers worldwide and approximately 162,700 employees as of March 31, 2013, Cognizant is a member of the NASDAQ-100, the S&P 500, the Forbes Global 2000, and the Fortune 500 and is ranked among the top performing and fastest growing companies in the world. Visit us online at www.cognizant.com or follow us on Twitter: Cognizant.

World Headquarters
500 Frank W. Burr Blvd. Teaneck, NJ 07666 USA Phone: +1 201 801 0233 Fax: +1 201 801 0243 Toll Free: +1 888 937 3277 Email: inquiry@cognizant.com

European Headquarters
1 Kingdom Street Paddington Central London W2 6BD Phone: +44 (0) 20 7297 7600 Fax: +44 (0) 20 7121 0102 Email: infouk@cognizant.com

India Operations Headquarters

#5/535, Old Mahabalipuram Road Okkiyam Pettai, Thoraipakkam Chennai, 600 096 India Phone: +91 (0) 44 4209 6000 Fax: +91 (0) 44 4209 6060 Email: inquiryindia@cognizant.com

Copyright 2013, Cognizant. All rights reserved. No part of this document may be reproduced, stored in a retrieval system, transmitted in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise, without the express written permission from Cognizant. The information contained herein is subject to change without notice. All other trademarks mentioned herein are the property of their respective owners.