The workflow before
EVERFI reaches more than 30 million learners, and corporate sponsors fund its courses in exchange for evidence of reach and outcomes. Its on-premise warehouse struggled with more than 100 million learning events a month. Reports took hours, new data sources took weeks to integrate, and the data team spent about 60% of its time on maintenance instead of analysis.
What we built
Over 32 weeks we moved EVERFI to Snowflake. Fivetran connectors now load learning management, enrollment, content interaction and sponsor reporting data, and Airflow orchestrates the pipelines.
We migrated the existing SQL logic to dbt models in bronze, silver and gold layers, refactored legacy SQL for Snowflake, and added dbt tests and documentation.
We rebuilt reports and sponsor dashboards in Tableau, opened self-service analytics to more than 30 business users, and cut over after a parallel run with minimal downtime.
Before and after
| Measure | Before | After |
|---|---|---|
| Report generation time | Hours | 90% less |
| Average query time | 30 seconds | 6 seconds |
| Data infrastructure cost | On-premise baseline | 40% lower |
| Pipeline maintenance time | About 60% of the data team's week | 60% less |
What's next
Tested, documented learning data in Snowflake is the foundation for Level 2 work, such as an assistant that answers sponsor questions about program reach and outcomes directly.
