Head of Data Governance at Northern Trust Corporation

When I was first asked to write this article, it was about the same time that ChatGTP went viral in the media and I must admit that I was intrigued. While excited to contribute content, like every data executive I am strapped for time so thought I would give ChatGTP a try. It also happened to be Women’s History Month so what better time to test and advocate for a product whose creation can be mostly attributed to their female CTO, Mira Murati.

I started with the basics by asking the chatbot “What is Data Governance?” and I received a solid, textbook answer not dissimilar from something that I have probably spouted off about in a meeting or presentation. “Data Governance is the overall management and control of an organization’s data assets. It encompasses the people, processes, and technology necessary to ensure that data is accurate, complete, consistent, and secure throughout its lifecycle, from creation or acquisition to deletion.” 

Data Governance typically comprises the following components:

1. Data quality: Ensuring that data is accurate, complete, and consistent

2. Data security: Protecting data from unauthorized access, use, disclosure, or modification

3. Data privacy: Ensuring that data is handled in accordance with applicable laws and regulations

4. Data architecture: Defining the structure and relationships of data within an organization

5. Data management: Managing the full lifecycle of data from creation to deletion

6. Data stewardship: Assigning responsibility for data to individuals or groups within the organization

7. Data policies and procedures: Establishing guidelines for the proper management and use of data within the organization

I opined on that statement and thought about the vastness of that mandate - especially at the financial institutions where I have spent most of my career. Where does one even start?

I have worn multiple hats in my tenure of working in the data field, but when you think about how data is the water piped through an organization and all the different areas that depend on it, it really comes down to one word – collaboration. There were times I made decisions independently that were less than celebrated because I didn’t first vet my decision through legal, compliance, information security, risk, and the list goes on. Even though they didn’t necessarily disagree with the decision, they still want to be consulted. The time it can take to get consensus across all these various areas is one of the main challenges with data governance and has often meant it’s often slow to show value. A data governance board/committee with the right decision makers actively participating and making key decisions can certainly help, but depending on the maturity of your organization will likely require business process reengineering and change management to effectively establish.   

“A data governance board/committee with the right decision makers actively participating and making key decisions can certainly help, but depending on the maturity of your organization will likely require business process reengineering and change management to effectively establish. “

Utilizing agile methodology can be another effective approach to data governance. Large, heavily regulated organizations still need policies, standards, and committees to manage their data and funnel decisions through, however, they need to move faster. I recommend leading with change management by influencing the culture to make it more data literate and savvy, then improving and implementing processes and lastly bringing in the tools or technology to support the journey. Incorporating the principles of agile software development to deliver value quickly, respond to change and collaborate closely with stakeholders will improve the speed and effectiveness of data governance initiatives.

This includes:

● Being iterative and incremental – Implement through a series of iterative and incremental steps, with each step building upon the prior one. This approach allows for continuous improvement and adaptation to changing business priorities.

● Being Collaborative – Agile data governance involves close collaboration between different data stakeholders, including business users, data analysts, IT professionals, and data stewards. It’s the glue that brings the ecosystem together.  

● Being Adaptive – Design your program to be flexible and adaptable so that it can respond quickly to changes in the data landscape. This may include adapting to changes in data quality, data usage, and regulatory requirements.  

● Being Value-driven – Focus on delivering business value quickly and continuously by aligning data governance activities to business strategy, goals, and objectives. Prioritize data governance activities based on their potential impact on business outcomes.   

● Being Data-driven – Monitor and improve data governance activities by establishing data quality metrics, conducting regular data quality assessments, and refining data governance processes based on feedback from stakeholders.

POCs can work well in data governance if you are just getting started.  I recommend standing up a small pilot group of business and technical data stewards to define data quality rules, document metadata, and establish an ongoing operating model focused on a key use case (such as; regulatory reporting, financial reporting, client reporting, or a critical operational process). They can also be early adopters of tools and technology to champion and support the overall program objectives.  

Every organization is unique and the ultimate objective of data governance is to ensure that data meets the needs of your organization’s stakeholders, data consumers, and end customers while managing risks and ensuring accuracy. Once implemented, effective data governance is an ongoing process that requires continuous monitoring, review, and improvement. Regular assessments of data management policies and practices can help organizations identify areas of improvement and adjust their strategies accordingly.