Data Governance as Usual
Congratulations! You've convinced your executives to support data governance. You've successfully completed that pilot project or addressed the first burning data platform. What next?
Creating a sustainable governance practice means transcending the pilot phase to become ‘business as usual'. Does your program have what it takes? Here are some indicators.
Your Program Has a Plan. And it's Business-Driven.
Keeping governance on the corporate agenda means keeping it warm with executives and stakeholders alike. Identifying information initiatives and requirements as part of—rather than separate from—established business planning processes ensures governance retains a seat at the table.
You Know When to Say When
Your governance program is complete when you have the minimum amount of data governance and management capabilities required to get the job done. No more. No less.
Frameworks and maturity models are excellent as a reference. Particularly when performing a health check or identifying areas for consideration. However, they shouldn't be used as a prescription or blueprint. One size does not fit all.
That being said: the job (i.e. your business objectives, data needs and pain points) will change over time. This is why your program needs a plan.
You Can Align With the Times
The composition of your data governance organization (or project team) reflects the nascent business priorities and organization dynamic at a point in time. But change is inevitable. Evolving business strategies, market pressures, or simply a change in leadership can signal the need for change. The composition of your executive steering committee, governance council(s) and stewardship teams must flex with the times. The operating framework may not.
It Doesn't Take a Village
The number of program participants and interested parties will grow with the scope of your program. Routinely gathering the full complement of council members or stewards can quickly become unwieldy and counterproductive.
To optimize productivity and minimize meeting overload, mature organizations adopt a working group mentality. In this operating model, participants—whether at a council or data stewardship level—are brought to bear on an issue based on the context of the data problem. Members not impacted or allocated decision rights need not participate.
Decision Making is "Baked In"
Ultimately, data awareness needs to be ‘baked into' day-to-day operational processes. What does this mean? Data management activities, including quality and risk assessments should be incorporated into software development lifecycles and production monitoring routines. This kind of control enables data issues to be avoided before expensive production problems—including compliance breaches - occur. It also ensures that needs for new data policies, standards and rules are identified preemptively.
In addition, operational decisions regarding data should be made as close as possible to the point and time at which data is being created and maintained. Clearly-defined decision boundaries and escalation criteria allow for intervention by stewardship teams, councils or executives on an as-needed, exception-only basis. This expedites issue resolution while promoting creation of clear operational data guidelines and data management practices. And, last but not least, it allows executives to focus on strategic rather than operational concerns.
Of course, baked-in decision making implies clarity regarding who does what when.
Can't answer the question of ‘what you (the data steward, developer, council member) do'? Start there.