Master Data Management (MDM) like many other “data terms” has many different meanings to different people, but most would agree that MDM refers to the processes and technical infrastructure that control the data assets of an Enterprise-Wide organization.
MDM is a set of processes, governance, policies, standards and tools used to manage the overall data assets of an organization. Management of the data includes, but is not limited to, processes for data collection, aggregation, matching, consolidating, distribution, synchronization, metadata, source identification, duplicate removal, application of standards, application of business rules, filtering, transformation, error detection and correction, mapping, enrichment and visualization.
One major goal of MDM is to have a single definition or version of truth for all informational entities that are used enterprise wide. Other goals may include unified data quality, metadata, lineage, data definitions, business rules, reusable access services (web services or Service Oriented Architect), changed data capture, data replication, synchronization, data governance, stewardship, and common access to disparate systems and technologies via “Federation.”
It is crucial that all MDM solutions be based on organizational processes and governance, irrelevant to the technologies and functionality used and is product agnostic. Equally important is keeping data and metadata (information about data) consistent throughout the organization, across every line of business, every business channel and application perspective.
MDM assumes a prerequisite level of maturity in Enterprise Data Management within the organization trying to implement it. A team of various architects are usually required to fully design and integrate an “end-to-end” MDM system that supports all of the “Enterprise Wide” business data requirements and the data flows between systems. It usually requires sophisticated software, hardware and techniques to make it all work together. The MDM infrastructure is composed of numerous technical roles, processes and technologies to support the data ecosystem of the organization. The infrastructure may use a number of different architectures, techniques and tools to support the organization such as HUB Architecture, Enterprise Information Integration (EII), Product Information Management (PIM), Customer Data Integration (CDI) and Master Data Integration (MDI) for integration, federation and virtualization.
While ideal, very few organizations have deployed a complete end-to-end MDM system and instead choose to focus on a subset of or specific area of MDM to deploy. Careful planning and design architecture must support both “Operational MDM” for OLAP and “Analytical MDM” for Analytical Systems such as “Business Intelligence” in the OLAP domain. There are four basic approaches to MDM implementation: 1) Consolidation approach 2) Registry approach 3) Coexistence approach and 4) Transactional HUB approach. Examples of such approaches can be found in product and customer data entities and may be addressed with several different MDM implementation approaches such as “Collaborate MDM,” “Customer Data Integration (CDI)” or “Product Information Management (PIM).”
Dataversal has designed and implemented a number of MDM initiatives over the years. We approach MDM as a business challenge and not a technology challenge as some others will make you believe. We start with the clients’ business plan and processes to better understand who uses what data, what is required by which business processes and how the data is used by each entity. A sound MDM strategy needs several other aspects to be in place including an established Data Governance and Data Stewardship model.
Dataversal has mastered the techniques to harness the power of this data and reporting technology for a competitive advantage in businesses of all sizes and industries. In addition to consultation, we offer a full range of Master Data Management Product solutions designed to lessen the experience level, cost and time required to deploy Master Data Management in your organization.