How do you maintain data governance skills in your analytics team?

I’m not sure there is one-size-fits-all data governance. While the tech might be a common denominator, governance-related (data) processes are specific to business needs. My experience tells that the best data governance experts are the one with a strong business experience.

An analytics team needs data, but that doesn't mean that the entire analytics team needs to be trained in data management principles. The analytics team should be on the receiving end of good data. They should understand where it is, what it means, and what they're allowed and not allowed to do with it. This is something that could be shared at the start of every analytics project.

Another common denominator shared between data analytics and data governance, is the necessity to adapt business culture into a data driven environment. Evaluation of your model maturity is a must, but contains risk of falling to the way side unless the necessary shifting in culture is accounted for.

Do not get hung up on titles. Data governance is not one thing, but rather is the strategic side of data management. Talk to information architects and enterprise architects, chief data officers, taxonomists and ontologists, business analysts, and CEOs. Most of those who hold governance authority have worked in the data field in a broad range of capacities. Finally, spend time talking with those who manage narrative content and social media, as these often play an oversized role in data aggregation and processing.

The analytical team needs to have a complete view of the company/institution's business process. It is important that the team can understand how it impacts business processes and what is the relevance of their work for improving the company's processes. Getting feedback from other teams on the progress achieved based on the results delivered by the analytical team is very important.

Data governance is increasingly about metadata. This information about information covers everything from classification to dimensional analysis, schema to constraint modeling. You will often deal with data from a wide range of sources, and this usually translates into subtle but important difference in definition. One key role of data governors is establishing what constitutes the ground truth when multiple languages (or ontologies) collide. Your curators and ontologists may be able to help identify how these languages, but ultimately it is your buy-off that establishes consistency within an organization.