Beyond data clutter: building the AI-ready data moat

Over the last decade, I have supported many teams as they navigated the shifting phases of data infrastructure. Looking back at the evolution of our industry, it is clear that we have moved through three distinct eras of friction:

• The 2010s (Data Silos): The challenge was simply connectivity—trying to link disparate departments via on-premises data warehousing.
• 2018 (Data Scatter): The shift to the cloud brought scale, but also fragmentation. We built data lakes and cloud warehouses that often lacked a central architectural "spine".
• 2022 (Data Clutter): We began searching for business value through concepts like Data Mesh and Data Contracts, trying to find order in the noise.

The 2026 Shift: The Data Moat

Today, the landscape is shifting again. As AI begins to commoditize code, the true competitive advantage is no longer the software, but the reliability and uniqueness of your data.

This is the era of the "data moat". A data moat ensures your models provide unique strategic value beyond what generic, public datasets can offer. It requires architecting the high-fidelity pipelines necessary to ground AI in proprietary, trusted data.

Bridging the Strategic Gap

I am now focusing my energy on this shift as a Fractional CTO.

My goal is to help growth-stage startups bridge the gap between their business strategy and the AI-ready infrastructure required to scale. It is a meaningful way to combine 15 years of technical depth at places like Roche, Equinor, and Tele2 with the leadership and mentorship I’ve developed along the way.

If your data currently feels like clutter, but your product needs a moat—let's discuss how to align your infrastructure with your growth objectives.