Save money and operational energy: State-aware orchestration in the dbt Fusion engine

"State-aware orchestration is a paradigm shift that changes what it means to build in dbt. Traditionally, a job builds everything that's selected. With state-aware orchestration, we only build the things that are going to change based on what we know about the project state." — Reuben McCreanor, Staff Product Manager at dbt Labs Reuben explains how state-aware orchestration in the dbt Fusion engine reduces both compute costs and operational overhead for data teams. Why state-aware orchestration matters: ✅ 10% cost savings immediately just from turning it on ✅ Build 20-30-40% fewer models by skipping unchanged data ✅ Reduces mental and operational costs for teams ✅ Only builds models when data or code state actually changes ✅ Simple to get started—turn on in staging, test, then production ✅ Fine-tune configurations for even more savings The problem: Across 33 billion models built with dbt, teams face two types of costs: - Actual compute and data warehouse credit costs - Mental and operational costs maintaining jobs and ensuring timely runs The solution: Traditional jobs are "dumb"—they build everything selected. State-aware orchestration is intelligent—it checks what's actually changed and only builds those models. Getting started: "As soon as you upgrade your project to Fusion, it's a simple matter of turning a staging environment onto Fusion, experimenting with state-aware orchestration, and then putting it on in production for immediate 10% cost savings." Reduce costs and operational overhead. Learn more about state-aware orchestration: https://docs.getdbt.com/docs/deploy/state-aware-about #dbtFusion #stateAwareOrchestration #dbt #costOptimization #analyticsEngineering #dataEngineering