Why Your Data Governance Program Isn’t Sustainable (and How to Fix It)
By now, many of the financial services institutions I work with have implemented made significant progress implementing data governance. But when I say “data governance,” I actually mean “data management.” For most, their data governance program is a series of disjointed processes held together by emails, spreadsheets and SharePoint sites. Does it work? Yes. Is it sustainable or efficient? Not at all.
So how do you make the transition from duct tape “data governance” to something real and sustainable? The answer is simple. Establish a system of record for data governance.
So I know what you’re thinking. You already have systems of record and authoritative data sources, and you don’t need yet another one. But I’d like to make the argument why you do. It comes down to making it easier for the organization to access trustworthy, easily-understood data more efficiently and consistently. In other words, reduce the time to takes to find data and increase the trustworthiness of reports: information governance.
Why Your Data Governance Program Needs a System of Record
First, a data governance system of record allows technical enablement of data governance processes. What that means is that you can create process flows that align to roles & responsibilities. You can also associate disparate governance processes (metadata, lineage, definitions, data quality rules, and results) with each other. And, you can create processes and workflows around data ownership and issue management.
Second, a data governance system of record also makes it easy for people to do the right thing. One of the most common arguments that I hear from customers is that their users can’t find the data that they need. And when they can’t find the data, they call Sally or Bob or whoever they think has the data. But Sally or Bob doesn’t always have the full story. And they may be providing the user with the wrong data for their need. On the other hand, a system of record, coupled with a data catalog, allows users to easily search for their data, add the data sets they want to their shopping basket, and checkout, all in one simple, automated process. And if they don’t know who owns the data, they can easily identify and contact that person through an automated workflow.
Finally, a data system of record makes it easy for data citizens to access the data you’re spending so much time governing. And when trustworthy data is accessible (and easy to understand), then great things happen. A data governance system of record links data lineage to metadata, to definitions and roles and responsibilities, and workflow. Now, data governance becomes integrated processes and workflows tied to the underlying data processes and systems. People can collaborate on decisions, close knowledge gaps, and improve the data. They can avoid data brawls – the situations we’ve all been in where everyone’s analysis of the same data returns different results.
Don’t Lose Sight of People and Process
Having a data governance system of record is a significant first step in turning data management into data and information governance. It provides the path of least resistance for data citizens to find, understand, and trust the data. But because data governance is as much as people and process as it is about technology, I also suggest that you do the following:
- Create an operating model that is easy to use and adopt for all your data user
- Clearly define roles and responsibilities as they relate to date, and make sure they are understood throughout your organization
- Establish authority – and not just a project sponsor. Make sure you have executives on board who can drive accountability for data governance within your organization. By having executives on board, you’ll have someone who can resolve escalated issues and decisions as well as define – and enforce – the consequences for non-compliance.
Wherever you are in your data governance journey, I encourage you to always look at taking one step forward. You don’t need to boil the ocean with every initiative or project, but look at ways to begin making that transition from data management to true data governance. And in the end, you’ll end up with a sustainable – and efficient – program that’s part of business as usual.