In a recent blog post, a Forrester analyst suggested that organizations should create a combined program for “information governance” that would cover governing both data and content. The premise is that there should be enough common ground and similarities in the two governance programs such that organizations can create “an information governance framework that will address both their structured and unstructured information [which] requires that the appropriate IT and business roles and responsibilities are clearly defined and that stakeholders from both IT and business are in agreement with the design and implementation efforts for an effective information governance strategy.”
This notion is a valid one – for example, why not look for where we already have roles defined and reuse those – however, there is a significant amount of work and consensus at such a high level that would need to be undertaken and successfully resolved to get to what Forrester suggests. It’s a fair bet that most organizations would spend so much time and effort – and management consulting fees – in getting agreement across both data and content silos and all the people involved, that there would be little to show for it from a business value point of view. They would run the risk of losing the interest and funding to continue.
Isn’t there a way to start small and grow?
I believe there is. Here’s a quick example from master data management. Many vendors and analysts are latching onto the “go governance, go early” mantra. If you have already started an MDM program, you may not have wittingly set up a top-down data-policy-led governance program to control it. So how do you “go governance early” if you’ve already started down the MDM path? There are a number of things you do as part of “master data governance” that can be the basis for jump-starting a broader data governance program, such as:
- Roles and responsibilities: you’ve defined data stewards and assigned decision rights
- Data sourcing: defined lineage, survivorship, enabled authoring/proposing new master data
- Set up business rules: defined data quality standards and data validation rules
- Governance processes: defined remediation and escalation workflows to resolve invalid data
- Audit and control: defined access rights and audit reports
- Publication: set up a process for deploying high quality master data to people and systems
Within this list are a set of governance processes that can be used as the basis for a broader data governance program definition. Even in the context of managing a data warehouse, many of these items are similar although they would be more focused on data integration, loading and results production processes rather than governance, audit and publication. But in either case, there is likely to be business value generated from your MDM and Data Warehouse program.
Raising up this level of activity to span two different but related initiatives as the Forrester article suggests is a noble goal, but don’t boil the ocean to get there. Look for quick wins by:
- Reviewing existing programs for governance activities you already do
- Document and manage data policies to formalize rules
- Use a policy-driven approach to formalize governance in other areas
You can think of this as a bottom-up way of getting a top-down data governance program started.