Getting Started with Data Governance in 4 Simple Steps
This week, I had the pleasure of attending the MDM/DG Summit in New York City. It was a great event, full of people looking to learn more about MDM and data governance. And what I found interesting is that so many companies are still at the early stages of their initiative, and are wondering how data governance can help their organization. And the question I heard again and again is “how do I go about getting started with data governance.”
Getting started with data governance can be a challenge. But I’ve found that the most successful data governance initiatives follow four simple steps.
1. Start with an operating model: successful governance initiatives start by creating an operating model that includes the required organizational structure, roles, workflows, and asset types. By asset type, I mean everything that is an asset to the organization, from reports and systems, to policies, databases, and data columns. Link the responsible roles and people to the asset types through a workflow process that defines how to handle changes in the available information. Collibra delivers an out-of-the-box operating model that help organizations get up and running quickly.
2. Contextualize the data dictionary and business glossary (and the link between them): many organizations already have a “data dictionary,” but without the link to a business glossary, it often misses the required context to makes sense of what the data represents. In addition, it’s very hard for business users to find the data as they are not aware of technical field names. In addition, I find that many organizations have a “business glossary,” but many times, it lives in Sharepoint or Excel. Linking the business glossary to the data dictionary is an essential step in enabling the business to find the data easily and ultimately take control of it. Also make sure that the business glossary contains all business-focused metadata (reports, business processes, and more).
Collibra helps organizations bring that data together through automated integrations, Excel uploads, and manual additions to the solution on an as-needed basis. Once ingested, your organization can now enrich and contextualize the data and establish the links between the data dictionary and business glossary.
3. Let the business lead: the most common scenario for a successful governance initiative is one where the business leads the effort. That way, the business is invested in the project, and is more likely to take ownership and drive adoption throughout the organization. Just to be clear, the most successful data governance initiatives do not just involve the business, but put them in the driver’s seat to lead the project.
4. Drive the project from the top down: by establishing what’s important at the top, your organization create a governance model that aligns with organizational priorities and expectations.
If you just scan all available data fields in the organization (often millions), it’s often overwhelming and difficult to put governance on all those assets.
For example, in the case of BI and analytics, identify the reports that are a top priority for the business. Then, pinpoint the key metrics on the report, and link them back to the data that is available across the organization. This approach helps you to identify the critical data elements, and establish a governance process around them.
Getting started with data governance is not easy. But following these four simple steps certainly sets you on a path for success.