A successful data mesh transformation is primarily an organizational journey, not a technical one. Building on last year’s presentation of SWICA’s data mesh platform, this session shares early lessons from the first months of a large-scale transition from a centralized data warehouse toward a decentralized data mesh. We focus on cultural, organizational, and operating-model challenges, and what they mean in practice across strategy, technology, and people change.

Target Audience: IT decision-makers, project managers, BI leaders
Prerequisites: Basic understanding of the Data Mesh concept
Level: Advanced

Extended Abstract:
Many organizations adopt data mesh principles with the expectation of faster delivery and stronger domain ownership, only to find that the hardest challenges arise outside of platform architecture. At SWICA, an initial phase of building the Dragonfly data mesh platform was followed by the start of a large-scale transformation program: the transition from a long-established, centralized data warehouse toward decentralized, mesh-based data products, with the goal of decommissioning the legacy warehouse within three years.

This session provides an experience-based view from the first six months of this transition. We describe how platform capabilities were combined with structured enablement of domain teams, including the onboarding of experienced practitioners to support domains during early migration phases. We also reflect on the organizational and management challenges encountered when initiating the program, including resistance, competing priorities, and the need for clear sponsorship and decision-making structures.

Rather than presenting a finished target state, we share early patterns, learnings, and trade-offs emerging from real execution. Attendees will gain practical insights into starting a data mesh migration, supporting domain teams at scale, and sustaining momentum beyond the initial platform build-out.