One of the LinkedIn Groups I participate in has spent the past two months discussing the following question: “Respect for data – are we beginning to see a shift from application centric to data centric enterprises?
One of the LinkedIn Groups I participate in has spent the past two months discussing the following question: “Respect for data – are we beginning to see a shift from application centric to data centric enterprises? Has the revolution started or am I just imagining?”
And finally – as such discussions often do – the topic veered toward the need for Data Governance. Person 1 suggested the need for a rule that would, in effect, force respect by requiring projects that impacted reference data to gain approval for their approach from a data-centric governance board. Person 2 concurred, adding ”Without Data Governance you have no controls in place. Without controls you’ve lost.”
This is such a common situation. Violent agreement for the need for governance, but not the meaning. Person 1 is looking at POLICIES, while Person 2 is embracing the need to enforce _policies_. Both are important, but any stakeholder who is just grasping the idea of making data-centric decisions rather than application-centric ones might be confused.
At the Data Governance Institute, we’ve started using the term “Big G” Governance to describe the policies, mandates, rulings, and rules of engagement that come from on high (wherever that is). An example is the rule that Person 1 suggests: Any project with a linkage to reference data is reviewed by a Board for a determination of whether the project is introducing data-related risk. Another such “Big G” Governance ruling might be that controlled reference data sets can be duplicated ONLY IF certain criteria are addressed that would ensure that the data stays in sync.
Of course, nobody wants bureaucracy, and “Big G” Governance never exists for its own sake. I had to agree with Person 2 in the discussion that data-related controls are what we’re ultimately aiming for. Controls need to be embedded in projects, processes, data flows, applications, and information management practices to ensure that the data and the people who touch it adhere to policies, standards, rulings, and rules of engagement. Here, down in the trenches, our objective is to institute a series of “little g” governance control points.
Of course, it can be tricky to translate policy to practice, so much of the work of Data Governance teams takes place in the all-important alignment layer between “Big G” and “little g” efforts. This is where stakeholders, subject matter experts, and experts get together to decide how to embed and enforce controls.
Application-centric or data-centric? The implementation of many “little g” governance efforts continue to be considered and managed from an application-centric point of view. That’s fine, as long as they’re the appropriate controls.
What I see changing, though, are awareness levels of mid-level and senior-level consumers of information.More and more of them seem to understand that:
- Inadequate Reports/BI/Analytics can result from inadequate “little g” governance controls.
- The people down in the trenches who work with those controls often get contradictory instructions, and that a key to aligned controls is aligned decision-making by managers and architects.
- Alignment activities require enforceable rulings and a certain level of empowerment to interpret them, embed them, and enforce them.
So these under-served stakeholders are calling out for data-centric “Big G” rules, rules of engagement, and councils to address gaps, overlaps, and conflicts.And – as Martha Stewart so famously says – “That’s a good thing.”
