3 powerful lessons of using data governance frameworks

The first published data governance framework was the work of Gwen Thomas, who founded the Data Governance Institute (DGI) and put her opus online in 2003. “Frameworks were already being used, but they weren’t publicly available,” she says. “I had been asked to help Coors Beer prepare for upcoming Sarbanes-Oxley audits. They already had a technical plan in place, and I helped them find the right size and structure of an accompanying data governance program. Throughout my time at Coors, I saw many examples of how they used the power of frameworks to keep everybody’s thoughts and actions in sync. That’s when I decided to write a more general framework that could be accessed by any organization and adapted to their needs.”

The DGI publication includes components it thinks should be included in a data governance program. And two decades after the first published data governance framework, a new version was put online. That first and only refresh was on May 3, 2023. In the meantime, organizations in different industries around the world have gained considerable experience using their own frameworks, which were often influenced by the one DGI originally shared with the world. It’s now clear that data governance is most successful when CIOs and CDOs do three things:

1. Involve all key stakeholders in the definition of a data governance framework. “You can’t assume data ownership is equivalent to the right to make decisions about the data,” says Thomas.
2. Start out with a clear idea of the business outcomes you want to achieve. “Focus on value,” she says. “Everything you do to collect, manage, and analyze your data ought to be traced to value.”
3. Use your framework to orchestrate execution. “Managing data and using data should be considered a portfolio of actions,” says Thomas. “When a good framework is defined, the CIO should be able to hand off tasks to different teams with full confidence not only that they’ll be performed accurately, but the outcome will contribute to achieving the overall goal.”

Who gets involved in defining frameworks?

The US Department of Commerce (DOC) is probably the biggest collector of data in the United States. They collect, archive, and analyze everything from weather and farming data to scientific and economic data.

According to Oliver Wise, CDO at the DOC, data collection is happening now for what is the most detailed collection of data on the state of American businesses. This survey is conducted every five years by the Census Bureau, just one of many agencies that make up the DOC.

“We ask detailed questions to find out what type of business they’re in, who their customers are and what their revenues are,” says Wise. “We find out about their employee base and whether they are contractors, part time, or anything else. These data provide a critical universal perspective on the state of the American economy that’s used by policy makers at all levels.”

Another important project currently underway at the DOC is the collection and analysis of data to inform supply-chain policy. The goal is to understand supply-chain choke points and predict them so the American economy can better react to shocks, such as those resulting from the recent pandemic.

In addition to the data it collects and generates from public sources, the DOC also buys or licenses data from the private sector and uses it for things like economic analysis. “The challenge is that when you take data from outside sources, you have to normalize the data to make sense of it,” says Wise.

Structuring the data and tracing the source are just two of many important aspects of data governance that are carefully considered by the DOC. The Data Governance Board, chaired by Wise, addresses data management and data policy issues for the wide range of agencies that make up the department.

“We have different data governance frameworks for different needs,” he says. “In all cases, the definition of any of the frameworks needs to be a collective effort, so all stakeholders feel they’re being heard. If you do that, everybody is more motivated to use the framework, which will ensure consistency in data management.”

What is the goal of data governance?

Hanna Hennig, CIO of Siemens, says she has seen business units start collecting data without knowing what to collect and why. “It was always a waste of money,” she says. “If you don’t know what problem you want to solve, then you cannot define your data strategy.”

To find out what data you need, start with a clear definition of what you consider to be the desired business outcome. Whether it affects the top-line, the bottom-line, or both, the desired business outcome will drive decisions about which data you collect. Once you identify the data, you can start defining your data governance framework.

The framework should answer questions, such as who owns each data asset, the role of the owner, and how you ensure the data is curated and qualified for use by the technology across the business. If the data is correctly curated and formatted, it can be used by data analytics and, in particular, AI to make recommendations that help an organization make decisions ahead of the market.

Poor data quality leads to poor decisions and recommendations. When the data is bad, you can’t make decisions ahead of the market and ahead of the competition—or worse, you make the wrong decisions. According to Hennig, data governance helps ensure data quality and prevent chaos in an organization.

“Without frameworks, people tend to protect their data,” she says. “When there’s no sharing, there are no use cases that span the value chain. If you’re not able to open data silos, you’re not able to harvest the benefits of the data across your company. The biggest value comes when you can implement end-to-end use cases—combining manufacturing with sales forecast planning, for example.”

Another important end-to-end use case is sustainability, which requires the first three scopes of greenhouse gas (GHG) reporting: scope 1 is on direct emissions from sources owned or controlled by an organization; scope 2 is on all indirect emissions resulting from an organization’s energy consumption; and scope 3 is an account of the emissions across the supply chain.

“All three require you to look across the value chain,” says Hennig. “Not only do you need to look at data within your company, but also outside your company, with suppliers and customers. You can’t do this if you have data silos.”

But most of all, Hennig says, organizations need to be clear about what problem they want to solve before setting up data governance. The goal should be to deliver business value.

How does your framework help teams work together?

Jennifer Trotsko, who founded the data governance function and later the privacy function at the International Finance Corporation (IFC), the private sector arm of the World Bank Group, was greatly influenced by Gwen Thomas’ work.

“We developed our own framework based on DGI components, along with other leading benchmarks,” says Trotsko, who went on to become the Head of IFC’s compliance risk function and chief privacy officer. With the foundation in place, IFC was able to coordinate activities across teams. And by using a common language to communicate about everything from policy and rules to technology and processes, each part of the organization could reference the framework and contribute to the overall end state.

“After establishing a project’s business value, the first thing we did was map tasks to our in-house data governance framework,” she says. “By assigning a lead in the policy area, another for working with technology, and another for change management, the project had clear guardrails and milestones. This allowed the core team to manage across dozens of departments, and it was the framework that provided confidence to stakeholders that all important components were covered. In short, leads focused on execution, knowing we all had a shared vision for the overall effort.”  

Trotsko is now privacy program manager at the International Monetary Fund (IMF), where she continues to build and adapt the DGI data governance framework, which she says is “invaluable in managing large projects involving data collection, storage, and analysis.”