Understanding the Basics of Data Mesh and its Impact on Data Governance

By Nicola Askham

In the rapidly evolving landscape of data management, a term has emerged that is simultaneously intriguing and confusing: Data Mesh. If you find yourself puzzled by this concept and its implications for, you're not alone.

Many of us in the data realm have been grappling with the question: What exactly is a data mesh, and how does it impact our approach to data governance?

Imagine encountering a new client who casually drops the bombshell that they're embarking on a Data Mesh journey and expect you to oversee Data Governance for it. Panic might set in, as you realize that while you've heard of Data Mesh, you're not entirely certain how it impacts Data Governance.

Data Mesh isn't just another technological marvel, like the migration of data to the cloud that prompted a flurry of questions about Data Governance changes a few years ago.

The Data Mesh concept encompasses more than a fresh technology stack or a novel infrastructure. It's about a distributed architecture that breaks away from the traditional data warehouse or lake model. Instead, it envisions data as a decentralised resource, accessible through various APIs and systems.

The crux lies in the shift in mindset that Data Mesh demands. It's not just about IT delivering solutions; it's a cultural change that invites all stakeholders to think differently about data ownership and accessibility.

While previous data warehouses and lakes could operate without airtight Data Governance (albeit suboptimally), the same isn't true for Data Mesh. It hinges on a cultural revolution where data becomes the shared asset of the entire business, requiring robust governance to maintain its integrity and usability.

The Democratisation of Data: Introducing Data Products

At the heart of Data Mesh lies a fundamental shift in how we perceive data's value and accessibility. The term "democratisation of data" is more than a catchy phrase; it's a philosophy that shapes how we approach data products.

Data products aren't massive data dumps; they're finely curated, bite-sized datasets that hold value on their own. These products are designed to be easily accessible and usable by a wide range of users across the organisation.

The concept of a data product may sound straightforward, but its implementation requires careful consideration. Not all data is meant to be a data product. The criteria for turning data into a data product hinge on its accessibility, understandability, discoverability, interoperability, and trustworthiness.

By adhering to these principles, organisations can ensure that their data products are valuable, usable, and ultimately contribute to the democratisation of data.

Adapting Data Governance for Data Mesh

As we explore the intricacies of Data Mesh, the question of Data Governance looms larger. How does Data Governance need to evolve to accommodate this new paradigm?

The first step is acknowledging that a one-size-fits-all Data Governance framework won't suffice. While a standardised framework can offer inspiration, each organisation's unique culture and challenges necessitate a tailored approach.

Roles and responsibilities play a pivotal role in Data Governance, and Data Mesh introduces some new players. The introduction of data product owners and data product development teams raises questions about the role of traditional data owners and data stewards.

The evolution of Data Governance in the Data Mesh era involves reconciling these roles, ensuring that data ownership and stewardship align with the demands of democratised, decentralised data.

In conclusion, the confluence of Data Mesh and Data Governance represents a transformational shift in how we manage and utilise data. Data Mesh isn't just about technology; it's a cultural and architectural evolution that necessitates rethinking our Data Governance strategies.

By embracing the democratisation of data and adapting our governance practices, we can navigate the complexities of Data Mesh and harness its potential for enhanced data usability and value.

Originally published on https://www.nicolaaskham.com/