AI is moving fast, but organizations don't transform just because the technology exists. Data leaders need to rethink how their teams create value, especially in the era of agentic AI. Success requires moving beyond isolated AI experiments to transforming end-to-end business processes—while acknowledging dependencies on legacy IT systems.

Key Points

• Agentic AI is process-centric: The real promise is transforming how companies work and create value, not just adding AI features.
• Legacy IT is the hidden hurdle: Existing core systems block AI's full potential and require modernization.
• Data teams cannot succeed alone: Tight alignment with IT and business leaders is essential for transformation.
• Revisit your operating model: Success requires reassessing value stories, reorganizing team structures, and operationalizing big moves.

Most AI initiatives in traditional industries add features without changing core processes. Agentic AI changes the conversation by putting business processes back at the center—combining LLMs with other technologies to transform and automate end-to-end workflows.

But legacy IT landscapes are full of outdated architectures, overlapping systems, and hard-to-integrate interfaces. Expecting agents to seamlessly navigate this complexity without significant modernization is unrealistic. Data and AI teams must accept their dependency on slower-moving IT foundations and work jointly on long-term change.

For 2025 and beyond, data and AI leaders should revisit their mission and operating model:

1. Reassess your value stories – Gain clarity on the tangible value your team delivers.
2. Relearn and reorganize – Observe how roles, skills, and expectations are shifting and adjust team structures.
3. Operationalize the operating model – Define and execute "no-regret" big moves for each value play.

Upleveling a data team for analytics and AI innovation is less about chasing the newest model and more about rethinking how the team fits into the broader organization. By focusing on processes, acknowledging dependencies on legacy systems, and deliberately reshaping the operating model, data and AI leaders can move from isolated experiments to meaningful, sustained business transformation.