In the years that I covered data governance and master data management (MDM) as a Gartner analyst, data architects and CIOs frequently asked, “Why is our data such a mess?”
Though I have written at length about how we got into this mess, I want to share how data governance and MDM are key to digital transformation and, more importantly, getting real business value from your data.
I often hear from companies that implementing MDM and DG presents a bit of a “chicken or egg” problem. They ask, “Should I first define my data governance and quality standards or work match, merge and clean my business data from disparate systems across the enterprise?”
Read on to learn how to best coordinate and implement these critical programs. But first, I need to put them both in the right business context.
The Business of Data Governance and Master Data Management
I argue that data governance and MDM are co-dependent disciplines. That is, companies cannot successfully implement one program without a commensurate focus on the other.
In this relationship, data governance is the “business function” of MDM, controlling how data is created, collected and used. For a governance program to reach its full potential of optimizing existing processes while generating new business opportunities, it needs to reside within the business as opposed to IT.
Once companies view their data governance initiative in this proper business context, they soon recognize the need for executive participation and stakeholder engagement. Then, this fully engaged group of stakeholders can formulate and ratify a set of business outcomes and engage IT to engineer a data model that services those outcomes.
Governing and Mastering the Most Critical Business Data
By first placing a data governance initiative in the proper business context, companies can focus on which data is most critical to their success and scope an implementation plan from there.
MDM implementers refer to this most critical data as master data, the core data within the enterprise that describes objects around which business is conducted. Master data is usually state-driven, slowly changing and shared widely across multiple source systems.
Just as key as having the business drive both governance and MDM, it is critical that that organizations properly scope their initiatives to avoid “boiling the ocean.” Governance, then, describes only the data entities and attributes that helps reach the previously defined business goals.
The key step here in coordinating MDM and data governance is to separate the target master data logically and physically from the non-master data in the model, as only the master data will be managed within the MDM system.
Determining Data Quality Rules and a Conflict Resolution Process
At this point, business stakeholders begin earnestly participating in data governance at all levels of the enterprise. At most organizations, leaders have formal meetings with their IT partners and begin the important work of governance, including:
- Resolving definitional conflicts between sources for each data element
- Recording any subsequent technical and business definitions, lineage and mapping or transformation rules
- Creating and documenting data quality rules
- Consolidating the number of definitions that are actually the same
- Adjusting the data model to reflect these governance rules.
At this stage, it is critical that business and IT stakeholders thoroughly examine their business data and co-author data definitions that help meet their organization’s goals.
Data Governance: An Application or a Set of Principles?
The most basic definition of data governance is that it is a set of principles and practices that ensure high quality through the complete lifecycle of an organization’s data.
Though it technically does not require a dedicated technology platform, data governance solutions have made it easier for organizations to scale their governance efforts and leverage their accumulated data assets in support of new business processes and analytics.
Some organizations acquire dedicated data governance or data catalog technology to track metadata, while others treat metadata as an additional master data domain and manage it within their MDM system.
Regardless of the technological approach, organizations need to walk through this “policy creation and management” phase of data governance to later track progress and resolve any further issues.
Once all conflicts have been resolved, the remaining metadata described previously can be captured, and the data quality rules also created and recorded.
Coordinating Governance and MDM: Monitoring, Execution and More
Policy creation and management is only the first step in implementing a data governance program and identifying the means to enforce its rules through an MDM implementation.
To learn more about the “monitoring and execution” phase of data governance — as well as tips on how to configure, load and implement the MDM data model — read the full article: “MDM And Data Governance: What Comes First, the Chicken or the Egg?”
With over twenty years of experience in the design, development, and implementation of large-scale Data Governance and Master Data Management (MDM) programs and systems, Bill O’Kane is VP and MDM Strategist at Profisee, a leader in Master Data Management (MDM) solutions. Bill served for eight years as Vice President of Data and Analytics and Magic Quadrant lead author at Gartner. He is a recognized thought leader and the world’s foremost authority on the topics of Information Governance and MDM program management, vendors and technologies, and MDM’s role within an overall enterprise information management strategy.