Master Data Management
SAP MDM provides typical features of Data Consolidation, Management and Data SAP MDM provides typical features of Data Consolidation, Management and Data.
However, sometimes SAP MDM is not capable of providing end-to-end Master Data Solution on its own. This paper explains how other technologies can be complemented with MDM to overcome the challenges and provide a complete solution.
SAP MDM (Master Data Management) supports Master data management and governance scenarios by feature like, import, syndication, workflow and matching and merging. Most of the typical master data management implementation needs are met with these out-of-the-box features. However, it falls short of meeting certain customer specific requirements of an end to end master data solution.
For a better explanation in this paper, let us define the terms MDM implementation and MDM Solution.
MDM Implementation: Implementation of Master data Management by using out of the box features of SAP MDM. This fits well in, green-field implementation of MDM which include establishing interfacing systems that provide the data to MDM and receive the data from MDNM Business Processes may need to be revised to suit features of SAP MDNM.
MDM solution : An end to end solution that satisfies customer-specific application logic around the master data before it is made available for operational usage. This fits well in, scenarios where Master Data Management is introduced into the already existing system landscape. And there is no possibility to revise the business processes to suit feature of SAP MDM.
SAP MDM Provides features which are required for an MDM Implementation, but fails to satisfy the needs of an MDM Solution. This paper outlines, with an example scenario, how features of CE and SAP PI can be complemented with MDM to provide an MDM Solution.
Architecture and benefits of MDM
Master data management is intended to bring a systematic approach to data integration that ensures consistent use and reuse of data. Customer data, particularly, is a concern, and this concern is aggravated by recent introduction of unstructured web activity data to a long list of data types found in customer profiles.
As big data architectures find greater use, the types of data in organizations grow haphazardly in structure, with the traditional hallmarks of poor data management – data duplication, incomplete data and error-filled records – continuing to be common. Master data managers set out to address these issues by, among other things, establishing a reliable data dictionary for use across systems and providing means to enforce standard.
As systems of record and customer engagement applications expand in use, companies find they have no single clear view of a customer. Transactional systems, analytical data warehouses and, more recently, interactional data from web activity provide different views of the customer, but ultimately, business users look for a single accurate view of that customer.
MDM systems typically include large repositories for master data storage. They also include change management and data pipelines, as well as workflow and collaboration facilities.
Along with customer-centric capabilities, MDM systems also include product, supplier and partner master data, with specialized master data systems targeted at special use cases, such as procurement or healthcare.
In recent years, aspects of MDM have come to be associated with metadata management, a related technology discipline that organizes descriptive “data about the data” in the organization. Also underlying some MDM implementations are data virtualization architectures that employ a data abstraction layer that enables data access without physical data movement.