Modernizing Data Delivery for Trust & Velocity
A real-world enterprise case study from EMC Insurance, where transparency, prioritization, and architectural modernization transformed a struggling program into a predictable engine of delivery.
Context
EMC’s enterprise modernization effort involved seven major business lines all transitioning off legacy systems at once. The stakes were high: underwriting, claims, billing, and analytics all depended on the same data platforms, which were now being rearchitected under tight deadlines.
When I stepped into my leadership role, the program was struggling with:
- Low stakeholder trust in data accuracy
- Manual, time-consuming QA cycles
- Teams working hard but not producing predictable outcomes
- A backlog of 1,000+ vaguely defined work items
- Cross-team dependencies causing constant rework
The symptoms looked chaotic, but the root issue was simple: the system did not support predictability. People were doing their best inside an environment that made clarity impossible.
Restoring Trust Through Systems Thinking
Rather than pushing harder or adding process, I focused on restructuring the ecosystem around clarity, truth, and shared ownership. This meant:
- Identifying friction instead of blaming individuals
- Mapping actual workflows (not theoretical ones)
- Building transparency into every step of delivery
- Introducing guardrails that supported teams rather than constrained them
This set the foundation for modernizing data delivery and rebuilding confidence across engineering, business stakeholders, and leadership.
Key Actions
- Modernized the enterprise data platform by transitioning off mainframes, implementing a lakehouse architecture, and adopting semi-structured models for consistency and speed.
- Defined critical fields, lineage, and operational metadata to strengthen data trust.
- Led cross-team alignment across seven business lines and 80–90 people.
- Integrated LLM-assisted metadata discovery to help teams query assets, terms, lineage, and quality rules using natural language.
- Established measurable QA signals using ABCs, regression standards, and automation scaffolding.
- Rebuilt prioritization to focus on the work that mattered most for the business.
- Replaced heroics with systems: clear handoffs, structured planning, and predictable delivery loops.
Transformational Outcomes
- Delivered four business lines concurrently in six months (vs. one every 9–12 months)
- Morale increased to 4.7 / 5 across teams
- Stakeholders reported a significant tr