Master Data Management: Considering the Customer First

Think of it as a highly accurate tool for making decisions and gaining insight about customers, products, and other aspects of a business with laserlike precision. Master data management (MDM) systems generate and maintain an enterprisewide system of record that contains consistent, reliable information necessary to perform vital business functions across a large organization. In the last few years, MDM has become an increasingly hot topic among enterprises due to its ability to help bring higher profits, improve customer service, increase revenue, reduce time to market, make it easier to comply with regulations, and enhance reporting and business intelligence.

Though experts agree that the holy grail of MDM is a solution that unites data from diverse applications (i.e., finance, billing, sales, R&D), systems, data warehouses, and locations, developing such an advanced IT architecture typically takes years, is extremely expensive, and virtually eliminates the chance for a fast ROI. A more reasonable approach is to implement a phased MDM strategy that enables enterprises to develop an effective working model for future development, while allowing for a fast ROI on the first phase of the project.

Knowing Where to Start
While businesses can choose from a number of data dimensions on which to focus, for many companies, customer data is a natural starting point for an incremental MDM approach. According to a 2006 research study by The Data Warehousing Institute, information about customers is the data most requested by users.

An incremental customer-centric approach to MDM can significantly enhance customer service efforts by providing more accurate and complete customer data and reduce the risk of human error. Access to accurate master customer data every time gives sales, marketing, and service teams a better vision of customer preferences and requirements, enabling them to provide targeted offerings and personalized service that improve customer relations. And because customer satisfaction can directly impact bottom line revenue, MDM solutions that are used to improve applications, such as CRM, can result in a fast ROI.

Three Implementation Solutions
Even though MDM is being recognized as a powerful way to solve data inconsistency issues, enterprises have been slow to adopt it because of its potential technology limitations, high costs, and extremely long deployment times. However, not all MDM solutions present such roadblocks. Whether implementing a comprehensive MDM strategy or a customer-centric approach, enterprises should consider one of three methods used for deployment: repository hub, registry model, or hybrid approach.

Repository hub--stores all master data in a single database, which provides always up to date and consistent master data. However, building a repository hub introduces a greater level of risk because this approach assigns the role of system of record to the repository hub, which is a role typically assumed out-of-the-box by many existing enterprise applications and legacy systems. Therefore, it is a difficult challenge to get repository hubs and legacy systems to peacefully coexist without creating an unstable environment. For this reason, building a repository hub is the most time-consuming and expensive option.

Registry model--allows master data to reside in the original application databases while the MDM hub creates and stores data keys that access and find all relevant records for a specific master data item. Compared to the repository hub, a registry solution is much more cost effective, can be more quickly deployed, and, if implemented one data source at a time, enables companies to achieve early ROIs, while migrating to a full-blown solution over time.

Hybrid approach--includes features of both the repository and registry methods. Like the registry approach, it leaves master data records in application databases, but also replicates the key attributes for each master entity in the MDM hub, which enables large distributed queries. Drawbacks of the hybrid method include the need to reconcile update conflicts between multiple application databases and replication latency issues.

Companies in the process of doing MDM due diligence would be well advised to pick a solution that is adaptable to the business processes and applications for which they want to leverage MDM and that has the flexibility to adopt all three models. Doing so provides companies with a maturity path for their MDM environments, allowing them to start with a registry model that provides the best time to value and then move towards a hybrid or repository hub model later if required.

The power of MDM derives from its ability to manage data across a company's entire information infrastructure. However, many companies lack the means or resolve to implement MDM at once across all applications, data sources, and physical locations in their enterprise. For these organizations, a phased approach is more cost effective. And for a good many of these companies, the first launching point for MDM should be the realm of customer service. Businesses that successfully implement a customer-centric MDM solution in one area of the company can then move on to other areas as business requirements demand, giving them the intelligence they need to improve business processes, build revenues, and increase competitiveness in the years ahead.

About the Author
Marty Moseley is CTO at Initiate Systems. Please visit www.initiatesystems.com.

Related Articles

MDM Inside the Box

Cognizant's modular master data management solution prompts industry pundits to ruminate about the space's maturity.

MDM in the Real World

Putting master data management into practice.

If MDM's So Great, Why Aren't Companies Mastering It?

Two recent reports throw a bit of cold water on master data management, despite significant benefits and business process gains.