Purpose
dbt Cloud is provided by dbt Labs, with the goal of helping data teams build reliable, governable data pipelines to accelerate analytics and AI projects. Fivetran and dbt have merged into one company. The platform is powered by the brand-new Fusion engine, giving data teams high productivity when building data models locally or in the cloud.
Key capabilities
- Transform data with SQL: Build data pipelines the way software engineers build applications—modular, tested, and version-controlled. Native SQL understanding and local validation can catch issues before data enters the warehouse, while multi-dialect compilation keeps logic portable as the data platform evolves.
- Interactive lineage (across projects and columns): See how data flows and transforms across projects through a real-time interactive DAG. Filter by model type, materialization, or dependency, and drill down to column-level lineage to trace where a field comes from, how it is transformed, or when it was renamed.
- Automatically refactor models and columns: Deep SQL awareness keeps projects consistent as they evolve. Rename a model or column once, and dbt automatically updates all downstream references; preview all proposed changes before committing.
- Interoperable by design: dbt does not store user data. The platform provides ecosystem integrations and follows open standards to move data securely, scalably, and flexibly from where it is to where it is needed.
- Intelligent orchestration: Fusion brings stateful intelligence to production pipelines, automatically building only the models that need updating, with no rewrites or complex setup. The company says this can save more than 30% on warehouse spend while maintaining SLAs and keeping projects running fast.
- IDE support: Get a native local development experience in commonly used IDEs, using IDE extensions for Cursor, Claude Code, Windsurf, or VS Code.
How to use
No specific installation commands or operating steps are provided in the material. From the official statements, users can build data models locally or in the cloud, write transformation logic in SQL, get metadata support through dbt Catalog and lineage, and perform local development in supported IDEs through extensions.
Use cases
- Improve data quality and trust: rich metadata drives dbt Catalog and lineage, improves governance, and provides trustworthy data for analytics and AI development.
- Increase efficiency and reduce costs: help teams identify inefficiencies, automate routine work, and use compute resources more intelligently, thereby reducing costs and accelerating delivery.
- Build better data for AI: ensure data is trustworthy, governable, and properly documented, so that AI is built on the best possible inputs.
Limitations
The material does not specify pricing, specific deployment architecture, the list of supported data warehouses, free quotas, or usage limits. The company emphasizes that dbt does not store user data; other details about platform compatibility and performance are not currently provided.
