2006 has been a big year for master data management (MDM). We thought we'd dedicate our year-end column to some of the lessons we've learned from our MDM customer projects.

Surprised by MDM's Early Adopters

When it comes to data integration, data warehouse teams got there first. Data warehouses leveraged business requirements, ETL (extract, transform and load) tools and data stewardship to define the rules, transform the data and populate platforms to manage historical data and support analytics.

Conversely, many operational systems still rely on point-to-point communication, employing different integration tactics and redefining data across projects. Few development teams apply corporate-wide standards for data integration. With each new project, they reinvent the wheel.

We figured data warehouse teams would sow the seeds for enterprise master data. However, MDM is also germinating in application development and technology architecture groups, where a smart individual or team realizes the promise of data integration at the operational level. Operational developers acknowledge the data integration prowess of their data warehousing brethren, but recognize that the data warehouse itself probably is not the answer. Because their needs are transactional, hygiene and standardization are more complex. IT managers on the operational side have discovered the promise of purpose-built customer data integration (CDI) and product information management (PIM) solutions, thus getting the MDM ball rolling.

The business intelligence (BI) community recognized that a common vocabulary was critical in communicating the results of operational reports and advanced analytics. However, most companies still lack standard terminology outside the data warehouse environment. Operational systems are frequently built with little awareness of company naming conventions and standards. This is because there isn't a true understanding of how heterogeneous the sources for master data have become.

Consequently, the number of master data sources continually grows. Each new custom-developed system or packaged application generates new data. "System of record," "system of reference" and "system of origin" arguments are intensifying. Successfully integrating data depends on the needs of the specific application and the rules around the content. Different development teams integrate data differently, according to the requirements of their systems.

This is why MDM is so critical - operational data access means centralizing and standardizing data so that it can represent the same business entity to different systems. For customer data, the beauty of CDI hubs is that they distribute the functional access but centralize the processing. Different operational systems have access to a single customer's name, but the cleansing, reconciliation and integration of that customer's data occurs in one place.

Data Governance Becomes Compulsory

Most companies have not yet mastered the handoff between business policy and data management. The premise of data governance is that business policies influence data implementation so that the data is stored, defined, protected and accessed appropriately and according to rigorously defined, enterprise-level guidelines.

Our BI projects taught us the necessity of information "tie-breaking." For MDM, such decision mechanisms are critical. Because it is less about centralized data and more about centralized data integration processing, there must be buy-in on who is permitted to define master data, how business rules are established and how inevitable conflicts are resolved. Not everyone should be allowed to change a customer's account or a product's price.

Centralizing access and business rules around master data means multiple systems sharing that data. So, conventions that establish which systems are allowed to access that data versus change the data versus delete the data should be formalized. Ignoring data governance risks customer attrition, revenue erosion, noncompliance and worse. These risks grow as data and systems proliferate.

Companies Launch MDM with CDI

We saw this coming as we were gathering case studies for our book, but it has been borne out since. Companies simply need flexible, real-time access to high-quality customer information. CDI has become the most common on-ramp for MDM.

CDI leverages a company's customer relationship management (CRM) investments. Sales force automation, campaign management, customer support and online systems can share the same data. The economies of scale are significant and the business opportunities abundant. Companies now know their customers at the time of interaction, whether at the point of sale, point of service or point of support. Executives understand the business value of integrated customer information.

After all, there is no such thing as a relationship with a product. CDI ensures CRM investments can be exploited by non-CRM applications, such as order processing and billing, across departments and geographies. It lets diverse systems and users see the same cleansed, standardized, reconciled and integrated version of an individual customer. Companies regain control of their customer asset. That is a problem solved, indeed!

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access