Open Framework, Information Management Strategy & Collaborative Governance | Data & Social Methodology - MIKE2.0 Methodology
Wiki Home
Collapse Expand Close

Collapse Expand Close

To join, please contact us.

Improve MIKE 2.0
Collapse Expand Close
Need somewhere to start? How about the most wanted pages; or the pages we know need more work; or even the stub that somebody else has started, but hasn't been able to finish. Or create a ticket for any issues you have found.

Exploring the Overlap between Data Governance Data Quality and Master Data Management

From MIKE2.0 Methodology

Jump to: navigation, search

Three terms that are sometimes perceived as synonymous, but really are different, yet fit together, are Data Governance, Data Quality and Master Data Management. I’d like to explain the overlap by starting with the most overarching concept, yet they are all overarching within an organization. They are like 3 co-centric circles.

It certainly can be true that the initiation of a Master Data Management (MDM) program can kick off the data governance and data quality programs. After all, MDM will be distributing its data to critical, and eventually if not right away, multiple, systems. This creates an elevated awareness of the data and raises the standard bar for data quality. It also implies shared contribution to the build of the golden records there, either by identifying an existing source or by building a governance workflow. This workflow would apply many of the rules that come from a data governance program.

Data governance, according to the Data Governance Institute, is the exercise of decision making and authority for data related matters. It’s the soft stuff that IS the hard stuff. It’s a program, perhaps supported by record-keeping software, but not MDM software. The program builds the rules for data quality for the enterprise. Of course, when we refer to the enterprise in an enterprise with MDM, we should be referring to MDM itself because that is where the enterprise will get the data.

Data quality should ideally be considered for its enterprise implications, which would mean data quality violations get addressed at the earliest point, which is often the point-of-entry. This is not always possible within timelines and domain of an MDM program. At least at the MDM level, the data will be cleansed. MDM data needs to be free of intolerable defects. The magnification of that data within the enterprise will be huge. This is often why companies shy away from MDM – the data will be spread and if it is wrong, then the enterprise will act on wrong information. However, with a good Data Quality program, directed by Data Governance, the MDM data will be clean according to defined enterprise rules.

In summary, each system has separate responsibilities for data, but the systems must work together. • Data Governance – make the data quality and process rules • Data Quality – enforce the data quality rules made by governance • MDM – distributed the data cleaned by Data Quality and implement process according to Data Governance rules

It follows that there should be some organizational cohesion between the programs. At some early point in the organizational structure, the programs should roll up. This is often under the MDM banner, which also means the programs would fall under the MDM budget. However, if data governance and enterprise data quality programs are established prior to MDM, as could very well be the case in a larger organization, these organizations could meet at the Vice President of Information Management, or equivalent, level.

The MDM program should look for leverage points for its data governance and data quality needs off of any existing program of this nature. After all, data governance and data quality are enterprise requirements and deliver enterprise assets. MDM is an important vessel for utilizing the assets, but the data warehouse, other analytic databases, operational business intelligence and various operational environments also benefit and/or are affected by these programs. Data quality, for example, should turn first to the operational environment to fix intolerable defects. And data governance will generally deploy to not only MDM, but also operational processes.

Clearly, MDM will need deployment resources. Data quality is implemented within MDM as well as the data warehouse, and potentially other programs, although the resources are often part of the program teams, especially in the data integration function. Data Governance is seldom a dedicated team and is most effective when its members have significant other business interests for the data. Establishing Data Governance, as well as all other business contribution, to an MDM program can be a challenge. However, some minimum level of contribution must be brokered. This virtual team, as well as all of an organization’s information management, needs executive oversight.

Organizations are increasingly sponsoring information at the executive level, though a step towards that that many organizations find themselves in is to have an Executive Sponsor for information projects. Ideally, this sponsorship is represented in the data governance process.

Wiki Contributors
Collapse Expand Close

View more contributors