• Christophe Barriolade

Meta Data vs Master/Reference Data, the missing link of data governance?

Updated: Jun 13, 2018

This week Orchestra Networks exhibited at the Data Governance Financial Services conference in New York and I had the opportunity to speak with data governance leaders from many global financial services organizations about the types of challenges they are facing in establishing effective governance programs.

One of the main discussion points with attendees stopping by our booth was the gap between "metadata" and "master/reference data". In other words, many companies have embarked in enterprise projects around those two topics and are now looking at a way to connect them.

If we look at metadata management initiatives first, we see a growing concern on how those business glossaries, enterprise data models, and data governance policies can be dynamically connected to enterprise operational data stores, in particular the MDM/RDM hub. While it might appear that this should be the obvious outcome of any project, in fact the normal state of affairs is that when a definition changes in a business glossary, no mechanisms are in place to ensure that such a change is automatically reflected in the systems that are intended to be aligned with the enterprise-wide dictionary.

When metadata management solutions are disconnected and have no a real connection with enterprise data stores, there is a risk of rejection by business teams.

On the other side, MDM/RDM teams struggle with metadata. While they try to implement semantically rich models into the MDM/RDM hub, they are now looking for a way to "Inherit" from the metadata management solutions so they can link their data model objects to the enterprise business glossary and also how to apply data governance policies (RACI) into the data management workflow and authoring front ends utilized by business users. Again, this reality shows that there is a disconnect between master/reference data and metadata.

A clear illustration of this disconnect was illustrated by Malcolm Chisholm in his session "Managing Core Reference Data in Finance" when he described the "false division for reference data" by drawing a wall between metadata and database design on one side, and data values in implemented databases.

The ability to provide a metadata layer dynamically connected to the MDM/RDM hub is a focus of Orchestra Networks. Connecting definitions of data assets, governance policies, AND the actual data content and authoring workflows in one unified solution both addresses end-to-end master/reference data life cycle management and raises the adoption of data governance across the enterprise. So stay tuned to future announcements!