Data Integration Roadmap Series - Part Two: Master Data Management

When planning your integrated enterprise data environment, it is impossible to understate the importance of master data management. Much has been written about MDM, and it encompasses a broad range of (mostly non-technical) disciplines that are beyond the scope of a single blog entry. Here we will provide a broad overview of the four main areas of MDM to start your journey towards enterprise data governance. We will also examine the relationship between MDM and recent developments around other enterprise management programs such as Product Information Management.

What is master data management? Quite simply it is the administrative oversight of organizational “data as an asset” to maintain its consistency and credibility. It is not dissimilar to the management of other assets (e.g. financial, human resources, etc.), and should be approached in a similar manner when determining budgets and priorities. Building a strategy around MDM most often means focusing on these four areas:

  • Data Quality

  • Data Integration

  • Data Governance

  • Metadata Management

Data Quality involves determining a set of rules around all elements of organizational data and insuring that such data is profiled and cleansed to conform to those rules. Data Integration (in the context of MDM) refers to the integrity of disparate but related enterprise data and efforts to create a unified view of that data. Data Governance entails the stewardship of operational and analytical data among key business stakeholders who are charged with overseeing its accuracy, veracity, stability and security. Metadata is a catalog of an organization’s proprietary nomenclature and business rules, and how those are mapped to the data itself.

Organizationally managing these four areas can be a challenge. There is often a perception that any kind of data management belongs in IT, but most of these disciplines require investment and oversight from the business. Most importantly, to integrate these efforts, there must be cooperation between IT and the business. For example, a metadata dictionary should be maintained and governed by the business, but it is usually IT that manages the technical “data dictionary”. Collaboration is needed on how the metadata can be tied to this data dictionary yet still owned and curated by the business stakeholders. Fortunately there are applications such as SAP Information Steward that can easily manage these tasks.

How does MDM fit in with an organization’s other information management programs such as Product Information Management (PIM)? For many companies, PIM programs most often focus on the customer experience at the end of the (usually digital) distribution channel. However, as these channels become more and more digitized, the data that these products and channels generate take on a larger role in management requirements. A complete MDM program should encompass the digitization of product and service distribution to provide a total strategic platform, rather than treating PIM as a separate entity (along with other enterprise management programs such as resource management and life-cycle management).

The first part of this series examined the role of the logical data model. When coupled with robust master data management, the enterprise now has a governed, structured foundation for their data assets. The next step is to create an analytical model to derive value from that foundation, which will be the subject of the third and final installment in the data integration roadmap.

About the Author

Joe Caparula is a Senior Consultant with Pandata Group where he guides clients on the modern data architecture framework.