From the course: Learning Data Analytics: 1 Foundations

Learning about data governance

From the course: Learning Data Analytics: 1 Foundations

Learning about data governance

“

- [Instructor] No two organizations are exactly the same, and certain organizations have to adhere to regulations and compliance. For example, public companies in the United States must be compliant with Sarbanes-Oxley, or SOX. In Canada, they have a version of this called C-SOX, And in Japan, it's called J-SOX. No matter the country or the name of it, it's all around compliance, and data is certainly part of the compliance process. I'm certainly not an expert on all things data governance, but I am a data analyst, and I'm impacted by the rules that are set by the governance plans. Whether or not an organization is required by law to follow certain data governance, they are likely to have a living, breathing data plan that has controls in place. I want to talk about these plans referred to as data governance and how it impacts you as an analyst. Data governance is more than just the actual data. It includes processes, technology, and people. Key goals for governance are trustworthy datasets that are understandable, correct, secure. And one of the key outcomes is high quality data. When exploring data governance inside an organization, one of the first areas to explore is the ownership of the data. Different areas of an organization collect different data. The way they collect it may be different from the other areas of the organization, and the softwares they use may be different. Here's an example. Salespeople in the organization, they use a software called CRM, or customer relationship management. Their data is used by other areas of the organization, but it's the sales group that have ownership of these data. The group that owns the data is likely to be involved in your access to that data, even if IT is the gatekeeper that establishes your permission to that data. The IT group or technology department will also have plans in place to secure the data, and they are required to maintain certain forms of compliance themselves. It's important to understand that the way you gain access to the data and what permissions you will be allowed is usually controlled by the data governance plan. In some organizations, only people in IT have access to the backend of all the data systems. They may or may not provide you some form of data warehouse to access to pull reports. Regardless of the systems in place or the people involved, if you have some form of data either directly or indirectly, you can build processes to work with that data. Keep great notes of where the data is coming from and how you gained access to it. You can always work within the constraints of the rules. It's just not always elegant. I support a company that, when we're working with their accounting systems, we discovered that because of the critical data in that system, the rules are you either have access to all of it, or you have access to none of it. By leveraging what we know about data, we're able to automate datasets so that people who need access to the data but not access to the system still have protected data. I found early in my career the more critical my reporting was to the decision-making, the more access I gained to data. Keep in mind, when you try to pick up the latest, greatest tools to deal with data, you really need a broad range of skills and tools because you don't always know exactly how you'll be working with any given dataset. It's also worth noting that organizations are sometimes early in their adoption of data governance. New companies, of course, are going to be new to data governance, and older companies that haven't been under high regulation are also still developing their data governance plans. To protect yourself and your organization, you do not need all permissions to all data. When I work with a company I only ask for access to the data that I need, and I only asked for read access. I don't need access to their entire data infrastructure. "And certainly, don't make me a global admin," the joke typically comes out of my mouth. Don't put me on the list of potential offenders, because if I don't need the access to the data because I'm not using it, then it's just a layer of protection for me and, well, for them. There's certainly more to learn about data governance, but one of the key outcomes is the quality of the data. We're all looking for high-quality data. That way, when we build reports or visuals, we can feel more confident about what we're reporting.

Contents