IntegrationAdvancedetlwarehousesnowflake
Implement ETL vs ELT patterns for Salesforce analytics warehouse
Real World Scenario
Analytics team wants Snowflake as enterprise warehouse feeding Tableau. Debate Fivetran ELT vs Informatica ETL transformations.
Expected Answer
• ELT: land raw Salesforce tables in Snowflake then transform with dbt
• ETL: transform in middleware before warehouse load for complex cleansing
• Use Bulk API or Fivetran connector for initial historical load
• CDC stream for incremental after baseline
• Soft delete handling IsDeleted=true records in warehouse models
• Data model flattening: polymorphic lookups and history tables
• SLA: freshness for operational reports vs weekly strategic models
Follow-Up Questions & Answers
Click to expand — each follow-up includes a direct, interview-ready answer
Main difference: use case and scale. ELT: land raw Salesforce tables in Snowflake then transform with dbt. ETL: transform in middleware before warehouse load for complex cleansing. Pick based on your integration pattern and team capability. ELT wins when warehouse team owns SQL transforms; ETL when compliance requires cleansing before landing. Optimize for scale and operational observability.
Architect Perspective
ELT wins when warehouse team owns SQL transforms; ETL when compliance requires cleansing before landing.