Accountable Data Governance
The term data governance is a very malleable term in practice. Organizations can differ dramatically in terms of what the data governance person or team does.
Too many are treating data governance as a mandate. It made sense to put this vital function in place, but then what? Will it maneuver within the organization without accountability? Without also establishing accountability and tangible delivery, establishing data governance is not helpful.
Those data governance organizations that are thriving define and execute on a charter that delivers to the organization both in support of projects and as a horizontal organizational function.
Successfully done, it is actually quite a facilitator role. The data governance function must form and rely on steering from a cross section of upper management, covering the range of the enterprise that data governance is meant to cover. This group provides strategic direction and arbitration of thorny, controversial data issues that may involve ownership, sourcing, quality, prioritization or data breaches.
Also critical to accountable data governance is the establishment of the data stewards, assigned on a subject area basis. The data stewards make important tactical decisions over their subject area including sourcing, transformations and quality, retention, support, data verification, business metadata, etc. Ideally, the data stewards report, at some level, to those on the data governance council.
All this governed data needs to go somewhere, right? Data Governance needs to strongly align itself with those entities that have high leverage in the organization. These include, if present, a master data management (MDM) hub(s) and the data warehouse(s). Of course, all data should be governed, but these vessels are efficient for the architecture and for governance. Data governance is an essential part of MDM in many other ways as well. In many organizations, a well-done MDM hub is the evidence of data governance effectiveness.
The data governors can then provide services to projects needing to use the master data. They do this by conveying an understanding of the data available and mapping it to the structures of the applications. There may also be data the application generates that should be master data. The data governors can do this reverse mapping as well.
Speaking of the master data, many organizations leave the data modeling to those performing it as a strictly technical function. Data modeling is not strictly a technical function and the data governors can be instrumental to effective data modeling for chosen data stores.
Also at the organizational policy-making level, there are global data retention policies, documenting data definitions, documenting governance decisions and providing access to data.
In reality, there is a lot of work for the data governance function, but it should be accountable and evidenced in the organization’s data quality.
This post was written as part of the IBM for Midsize Business program, which provides midsize businesses with the tools, expertise and solutions they need to become engines of a smarter planet. I’ve been compensated to contribute to this program, but the opinions expressed in this post are my own and don’t necessarily represent IBM’s positions, strategies or opinions.