Blog

When I first heard about Data Mesh, I thought it was just another buzzword. But the more I explored it, the more it challenged how I think about data ownership. Yes, Data Mesh may be a bit of a utopia, but that doesn’t mean we can’t adopt its best principles to improve how we work with data today.

What is Data Mesh?

Traditional data governance, where one central team controls all the rules, often can’t keep up with the fast pace of business. That’s where Data Mesh comes in, a new way of thinking about data. It was introduced in 2019 by Zhamak Dehghani. The main idea? Treat data like a product and let the teams who know the data best take care of it. These are often the domain or business teams, like supply chain, sales or finance.

Instead of relying on a central data team, typically part of IT, to manage everything, each business team takes responsibility for its own data. This ensures the data is high quality and easy to find. This is called decentralized ownership.

At the same time, data governance provides a common framework to keep things organized. It defines rules around data quality and security while allowing business teams to manage their own data within those guidelines. Having a central data platform, such as Microsoft Fabric, plays a key role by providing the tools and infrastructure to support both governance and decentralized ownership.

What works well in practice?

1. Start small Don’t try to apply Data Mesh everywhere at once. Begin with 1 or 2 teams that are motivated and have good data skills. Use them as a pilot.
2. Treat data as your product Encourage teams to think about who will use their data and how. Like any good product, data should be of good quality.
3. Give business teams the right tools Data Mesh works only if teams have the tools and knowledge to manage data themselves. This may include training and support from a central data platform team.
4. Set clear standards Even with decentralization, you still need common rules for data definitions, quality, and compliance. A data governance framework helps maintain trust and consistency across the organization.

The Data Mesh utopia: why the ideal is hard to reach?

1. Full decentralization It sounds great to give every team full control of their data, but not all teams are ready. Some don’t have the time, skills, or interest. So: start small!
2. Perfect data products In reality, data products often lack good documentation or aren’t updated regularly. It takes effort and a mindset shift to treat data like a real product. Improving this is an iterative process. Business teams need to start small, define ownership, and slowly build habits around maintenance and documentation.
3. Complete culture change Data Mesh is not just about technology; it’s a cultural shift. Teams must work together and treat data as a critical asset that needs ownership. But this shift is hard. It challenges deeply rooted habits and assumptions. Shifting to shared responsibility means changing roles, breaking silos, and building trust between teams. That takes time and clear leadership.

Data Mesh as a journey

Data Mesh is not magic. The goal is to make progress, not to reach a perfect utopia. If you apply its principles step by step, Data Mesh can make your data strategy less IT-driven and more business-focused.

Think of Data Mesh as a journey, as in the picture below. It highlights three key principles: clear data ownership, treating data as a product, and ensuring accessibility. Progress on these three principles strengthens your data governance over time.

Author: Gert Jan Toering (Data Strategist)

At Chain Analytics we help businesses to enable the Data Mesh principles, helping them enhance data-driven decision-making. Feel free to reach out to us if you would like to know more! https://chainanalytics.io/contacts/