How EF Education First built a cost and value layer for dbt models with a Snowflake Cortex Agent

At dbt Summit 2026, EF Education First’s data platform lead Adrien Ledoux described how the company used dbt, Snowflake account-usage data, Power BI activity logs, and a Snowflake Cortex Agent to create a cost and value scoring system for decentralized data teams. After moving from fixed-cost SQL Server infrastructure to Snowflake’s consumption-based model, EF faced growing costs, underused pipelines, and unclear ownership of expensive workloads. The system evaluates dbt models and tests by execution time, credits consumed, and processing efficiency, then compares them with similarly sized models and classifies them into tiers that guide review and optimization efforts. The initiative improved efficiency on some major pipelines by 40 percent, reduced one model’s runtime from 10 minutes to 2 minutes, and led to the removal of up to 70 percent of reports in certain low-use Power BI workspaces. EF is expanding the approach through reverse ETL, Salesforce assignments, Confluence documentation, Jira automation, and usage digests, while emphasizing that usage alone does not establish business value and that adoption requires sustained change-management efforts.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.