dbt
FreemiumTransform data in your warehouse using SQL and software engineering best practices.
Published 29 May 2026 · Last updated 27 September 2026
Scores
Popularity3/5
De facto standard for data transformation in the modern data stack.
Learning Curve3/5
SQL-first approach is familiar, but project structure, refs, macros, and tests take time.
Flexibility4/5
Macros, custom tests, and the package ecosystem make SQL transformations highly composable.
Performance3/5
Transformation speed depends on the underlying warehouse; dbt itself adds minimal overhead.
Portability4/5
SQL-based transformations are relatively portable; moving to SQLMesh is feasible.
About dbt
dbt (data build tool) is the transformation layer of the modern data stack. Analysts and engineers write models as SQL SELECT statements (or Python on supported warehouses), and dbt compiles them, works out the dependency graph between models, and builds them as tables and views inside the warehouse. Adapters connect it to Snowflake, BigQuery, Databricks, Redshift, PostgreSQL, and other platforms.
The point of dbt is bringing software engineering practice to analytics: models live in Git, Jinja macros make SQL reusable, tests assert things like uniqueness and referential integrity, and documentation and lineage are generated from the project itself. Packages from dbt Hub add shared macros and models for common sources.
dbt Core is the open-source CLI under the Apache 2.0 license; its next generation, dbt Core v2.0, is built on the Fusion engine, a Rust rewrite that parses large projects far faster. The hosted dbt platform (formerly dbt Cloud) adds a browser IDE, job scheduling, CI, a catalog with lineage, and the Semantic Layer for governed metrics, with a free single-developer plan and per-seat paid plans. dbt Labs merged with Fivetran in 2026, bringing ingestion and transformation under one company.
Key Features
- SQL-based transformations with Jinja templating
- Version control and CI/CD integration
- Automated testing and documentation
- Modular and reusable model architecture
- Dependency management between models
- Adapters for all major data warehouses
- Open-source CLI (dbt Core) or the hosted dbt platform
- Semantic Layer for governed metrics on the dbt platform
Pros
- Uses the SQL analysts already know, with Jinja for reuse
- Git, code review, and CI apply to analytics code
- Tests, docs, and lineage are generated from the project itself
- Core is open source (Apache 2.0), with a hosted platform available
- dbt Hub packages cover common sources and patterns
- Adapters for every major warehouse
Cons
- SQL-only approach can be limiting for complex transformations
- Can become difficult to manage at very large scale without careful organization
- Testing capabilities are less sophisticated than traditional software testing
- Performance depends on underlying data warehouse
- Steep learning curve for Jinja templating for non-programmers
dbt Pricing
Freemium- · Free CLI, Apache 2.0
- · All transformation, testing, and docs features
- · Community support
- · 1 developer seat
- · 3,000 successful model builds a month
- · Browser IDE, job scheduling, and CI checks
- · Per seat per month, up to 5 developer seats
- · 15,000 successful model builds a month
- · dbt Catalog lineage, Semantic Layer, and API access
- · Custom seat count and 30 projects
- · 100,000 successful model builds a month
- · Column-level lineage, dbt Mesh, SSO, and RBAC
- · Contact sales for pricing
- · Everything in Enterprise, with unlimited projects
- · PrivateLink, IP restrictions, and hybrid projects
- · Contact sales for pricing
Tech Stacks with dbt
MLOps Pipeline
ProjectProduction-grade ML infrastructure. PyTorch for model training, Apache Airflow (or Dagster or Prefect) for orchestration, dbt for feature transformations, and Snowflake as the data warehouse, with Docker as an optional containerization addition.
Backend
Programming
Databases
Hosting
Power BI + dbt + PostgreSQL
ProjectStore data in PostgreSQL, use dbt to build clean and tested transformation models on top of it, then connect Power BI directly to the transformed layer for self-service reporting.
Airbyte + dbt + Snowflake + Tableau
ProjectAirbyte loads raw data from 300+ sources into Snowflake; dbt transforms it into documented, tested models; Tableau connects directly to Snowflake for governed self-service analytics and executive dashboards.
Tools Related to dbt
Works well with dbt(2)
dbt-clickhouse runs transformation models directly on ClickHouse — the dbt adapter supports incremental models, snapshots, and distributed table strategies.
dbt transforms the raw data that Airbyte loaded — the canonical ELT pairing for the Modern Data Stack.
Integrates with dbt(11)
Snowflake is dbt's most popular cloud adapter — full support including incremental models, snapshots, and dbt Cloud.
BigQuery is one of dbt's primary supported adapters — dbt-bigquery runs transformations directly in BigQuery.
dbt-redshift is an official dbt adapter with full support for Redshift SQL dialect.
dbt-postgres is dbt's original adapter — first-class support with all dbt features.
dbt-duckdb adapter runs transformations locally in DuckDB — widely used for development and testing.
Fivetran runs dbt models after each sync — the standard ELT pipeline (Fivetran loads, dbt transforms).
Vendor
dbt Labs
Website →Tags
Details
- Maintained
- Yes
- Tool type
- Transformation
- Primary language
- SQL
- Hosting
- Cloud & Self-hosted
- Open source
- Yes
- GitHub stars
- 13.9k
- Stars updated
- 2026-09-23