by Christophe Barriolade, CEO Orchestra Networks

If Master Data are the DNA of your business, MDM with Data Governance is its genetic engineering. 

Why Governance of Reference and Master Data must be addressed as a Pro-Active Business Initiative and should not be considered as a curative technology project .

Reference and master data are the DNA of any organization. They define all the facets of your business and reflect the value and differentiators you provide within the market. Products, customers, channels, locations,  geographies, accounts, organization, employees, suppliers etc., are the critical assets at the heart of your business. 

The definition of DNA on Wikipedia can easily be applied to reference data: "(DNA) is a nucleic acid that contains the genetic instructions used in the development and functioning of all known living organisms and some viruses. The main role of DNA molecules is the long-term storage of information. DNA is often compared to a set of blueprints or a recipe, or a code, since it contains the instructions needed to construct other components of cells." 

As with DNA, reference and master data are the codification of your business, shared across all business lines, consumed by all IT systems.  

So, if reference and master data are the DNA of your business, data governance is the genetic engineering. It means that the main purpose of a data governance initiative is to improve the quality, consistency and relevance of this data across the entire organization, not to fix issues after they occurred.   

As in biology, improving the  quality of your data cannot rely only on curative techniques. While data quality and data integration solutions are a key foundation for cleansing and connecting your data, you need to provide your business users with an active control  on their shared data. Data governance is a pro-active business initiative that has a real benefit to enabling efficient and effective business initiatives or compliance requirements. 

Semantic data modeling and Model-driven MDM  

If your goal is to gain a real control on your data, you cannot avoid the data modeling exercise. Without a common and unified description of your data, how could business users share the same concepts? 

In this book, Pierre Bonnet introduces the concept of a Model-driven MDM based on semantic data modeling.  Far beyond traditional models, semantic models describe your data in meaningful terms for all stakeholders, including business users. This means you can design a rich description of your reference and master data and hide or bypass the usual constraints of IT relational oriented modeling such as join tables or frozen cardinality links. 

Then it becomes possible to define complex data objects, mix hierarchical, relational and object-oriented concepts, configure business rules and validation controls, add documentation, etc. 

Semantic data modeling associated with the Model-driven MDM allows business users to be involved from day one in your data governance program. With a model-driven solution they can easily collaborate on data modeling and quickly achieve a description of their data in their shared business language because "what you model is what you get".  

While data modeling requires effort, the realization of a mutual and shared understanding for the whole business will become of recurring importance for a pro-active data governance program. It is the first step to building the best version of the truth and establishing a unique reference and master data repository with active data governance capabilities. 

Power to business users 

Once data models have been designed, your data governance journey is not over. Building a unique description of your data is useless if business users cannot gain control on data itself. This means that to be active, an MDM/data governance solution must provide not only a central repository for storing the truth, but also a full set of data management features and a user experience for collaboration that maximizes adoption.

Pierre Bonnet proposes an exhaustive description of the core capabilities that business  teams need to apply for pro-active governance on their data. 

It starts with a rich user experience in order to provide data owners, stewards and managers with a collaborative environment for managing data and improving quality overtime. It also addresses key issues such as version control, security, business processes and rules and finally integration of master data across information systems. 

Based on his extensive experiences at Orchestra Networks but also the MDM Alliance Group and Sustainable IT architecture communities, Pierre proposes an unbiased perspective on MDM/data governance methodology, in order to help you build a truly pro-active data governance program.