Almost seven years after Zhamak Dehghani introduced the concept of “data mesh,” it continues to be one of the most significant and controversial topics in our industry. More than just a data platform architecture, it’s more accurate to describe it as a socio-technical paradigm that originated in response to a critical question plaguing organizations: How can we efficiently and responsibly drive value from data at scale, across the entire organization, without grinding to a halt?
Organizations that attempted the journey and achieved success discovered that it is a complex transformation that takes time to complete. However, data-driven businesses are relying on a few fundamental principles, particularly the idea of data products. Whether one wants to even embark on such a journey remains a question of willingness to change.
In 2025, investment in data and AI grew significantly. The C-suite imperative to become “data-driven” has never been stronger. However, for many, this is still an aspirational goal that is always just out of reach. We’ve seen high-profile success stories from digital natives and bold incumbents, but we’ve also seen a quiet graveyard of stalled projects and failed implementations. The community is thousands strong, but the unclear boundary between the hype of a new technical architecture and the messy reality of its implementation persists. We also still see a profound disconnect between business strategy and data initiatives.Data teams, rich with talent yet siloed in IT cost centers are often limited by project-based, short-term budget cycles that are fundamentally antithetical to the long-term, product-centric model data mesh requires.
