For any data leader taking on a new project, the quality and security of data are always top of mind. Data governance is the key to ensuring that the goals of the organization can truly be met via the implementation of data. 

Data governance gets a bad rap because it’s often presented as bureaucratic or even punitive given the breadth of policies, processes, roles, and standards required. In my experience, building a data governance framework is a vital investment that pays short- and long-term dividends, as long as it’s approached with the right frame of mind.

Data governance also plays a critical role in answering some of the most important questions in companies today:

  • How can data work with maximum efficiency for teams operating in cloud or multi-cloud environments? 
  • How do you balance getting colleagues the resources they need to build literacy and effectiveness while ensuring they’re not spinning up new sources of (often duplicative) data that could be adding substantial costs? 
  • How do you create standardized approaches and appropriate views of data that have everyone’s buy-in because they’re built on a collaborative process? 

Here are five practices I’ll be following as I settle into my new role leading data analytics. I offer them up for consideration to any data leader that might be interested in how to create a data governance framework.

  1. Take a business problem perspective. What is your company trying to understand with its data? Which business goals revealed the need for a data governance program? It falls under the data leaders roles and responsibilities to ask questions like these that will help define and guide a data governance strategy and roadmap. It’s also up to you to bring in representatives of the business to provide input on what they’re driving toward as well as the data quality and views of data they need to perform at their best. Business users will come up with scenarios you’ve never considered. It’s your job to take care of the rules to make those scenarios possible, and to continue to have conversations with business users as those needs change.
     
  2. Articulate a vision while framing data governance as a dynamic journey rather than a static destination. Traditional data governance has been presented as a series of tasks that lead to a final destination. We do it, we do it, we do it, and then okay, it’s done. But with so much change happening in the world around us these days, having data that’s up to date and accurate argues for a data governance that’s more like an ongoing journey. Think of it in these terms as you shape the overall vision for where you’re headed. Yes, it may take you a few places you might not have expected, but it’s going to be great fun doing it. 
     
  3. Identify reasonable program milestones that build momentum. If you start a marathon solely focused on the finish line, you might intimidate yourself into not starting at all. Get your team to think in terms of clear upcoming milestones that you define together, and that get you from stage to stage, and you’ll see more efficient progress.
     
  4. Keep communications simple and clear for all stakeholders. The best way to tackle most big problems is to break them down into smaller constituent parts, and this is doubly true for how you communicate about data governance. Like it or not, you will have to convince people in key decision-making roles of what they're going to have to take on and what they'll have to relinquish, so clarity and directness are key. It’s also easy to overwhelm people when you talk about things like metadata, master data, and data architecture, so focus on keeping your explanations simple. 
     
  5. Create transparency about the challenges as well as the opportunities. A data governance model that defines any corporate data initiative as a black box – where you can see only inputs and outputs but no internal workings – is a self-defeating impulse. If everyone in your company has the potential to be a data leader, why would you use anything but transparency and openness in a data governance initiative? Transparency is without question the fastest way to build trust..


Guardrails over gatekeepers
Gatekeepers sound old-fashioned in organizations today, and data governance should be no exception. It’s in any data leader’s best interests to enable success by getting users the data they need rather than holding data close to the vest. I’m a fan of saying, “Right. Here are the guardrails. If you want to do your thing, great, but these are the guardrails. You can go off the guardrails, but you will be responsible for the consequences."

Framing data governance as a way to make better decisions, grow data literacy, and enable business success are the best ways I know to build acceptance, and dare I say it, enthusiasm.