Optimizing your runs for lower compute, fresher data, and faster iteration with dbt State

dbt State is a feature designed to reduce unnecessary warehouse computation by allowing dbt to remember prior runs and rebuild only models whose code, dependencies, or data have meaningfully changed. It supports model-level freshness policies through configurable lag tolerances and definitions of new source data, helping teams replace complex schedule-driven workflows with fewer, more targeted jobs. In development, automatic production cloning and deferral let engineers or AI agents test individual models without rebuilding upstream dependencies or manually handling manifest files. A demonstration showed capabilities including skipped unchanged models, manifest-free state selection, zero-copy table cloning, explanations for execution decisions, and reuse of teammates’ development outputs. Fanatics reported 25–30% model reuse and about 15% Snowflake cost savings in a pilot. Planned additions include query-history-based freshness recommendations, column-aware change detection, and a UI for comparing changes, while source freshness alerting, changed views, and snapshots remain outside or limited within State’s scope. dbt State is available as an installable option for dbt versions 1.7 through 1.12, is integrated into dbt v2, and is enabled through the dbt platform, with usage-based pricing tied to daily target table reuse.

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.