Customer Story

Operationalizing AI at Scale: How MCP is Driving ROI at EMC Insurance

What we did

Implemented an AI framework that enabled the client to rapidly stand up MCP for 3 major use cases: automate lineage discovery, accelerate code migration into dbt, and analyze Power BI semantic models in minutes instead of months.

Reduced multi-quarter engineering efforts to weeks, surfaced hidden metric definitions across 100,000+ SQL extracts and BI models, and created a foundation for EMC’s long-term AI initiatives

Industry

  • Insurance

Technologies

  • dbt
  • ·
  • Power BI

The Challenge

As EMC Insurance, a $2 billion property and casualty insurer, set out on a company-wide data modernization effort, the biggest challenge became scale. A large, complex legacy environment made it difficult to move quickly while maintaining consistency in how key business metrics were defined and used across teams.

With dbt identified as a key enabler, EMC needed a faster way to transition a large, complex data environment while gaining clarity into how critical measures were defined and used across teams. The goal was to reduce manual effort, increase confidence in reporting, and free analysts to focus on higher-value work.

  • “There’s been value in every use case we’ve done. But if you can find a few slam dunks like this, it [MCP] makes everything worthwhile.“

    Rob Vicker, Data Architecture Director at EMC

The Solution

EMC partnered with Analytics8 to accelerate migration work, surface business logic buried across systems, and give teams faster, clearer insight into how their data was being used. To accomplish this, we:

  • Documented new models, definitions, and business rules for consistent adoption across the enterprise.
  • Deployed an MCP framework that enabled EMC to analyze warehouse objects, legacy SQL, and Power BI models through natural language, giving teams immediate visibility into definitions, dependencies, and data logic that were previously difficult to trace manually.
  • Accelerated the transition from SQL-based views to dbt by using MCP-assisted parsing to convert existing code into dbt models, allowing EMC to replace months of manual effort with an automated, repeatable process guided by Analytics8’s expertise in modern data engineering patterns.
  • Leveraged MCP to identify every version of key business metrics across SQL extracts and Power BI, providing EMC with a clear understanding of how each measure was calculated and the insight needed to move toward one governed, trusted definition.
  • Incorporated MCP-built tools to inspect Power BI semantic models so teams could quickly see how metric logic was constructed, where differences existed across reports, and how those inconsistencies affected downstream analysis.

The Results

EMC partnered with Analytics8 to accelerate migration work, surface business logic buried across “There’s been value in every use case we’ve done. But if you can find a few slam dunks like this, it [MCP] makes everything worthwhile.” – Rob Vicker, Data Architecture Director, EMC Insurance

By combining MCP-driven automation with targeted process improvements, we helped EMC move quickly toward a modern, governed data foundation that supports consistent reporting and future AI initiatives.

  • Migrated 130 SQL-based views to ~1,000 dbt models in 4 weeks, reducing a previous year-long ETL migration effort to just 1 month.
  • Power BI model analysis that previously took months was completed in days, giving teams a clearer understanding of how core business metrics were defined across the company.
  • Redirected $6 million spent annually on manual metrics reconciliation towards strategy that supports the company’s broader modernization and AI goals.
  • Enabled data discovery and lineage across 45,000+ EDW objects, giving teams access to logic and metadata in under a minute per query.

More stories like this

Related blogs

Talk with a data expert

Schedule a meeting with one of our data analytics experts who will review your data struggles and map out steps to achieve data-driven decision making.

Request a session