dbt Semantic Layer

Overview

dbt Semantic Layer puts shared metric definitions on dbt models and serves them to analytics tools, applications and AI systems. It is best suited to data teams already building with dbt that need consistent, governed measures across several destinations. The central trade-off is that using the layer requires a Starter or Enterprise account, with monthly query quotas on every qualifying plan.

Key features

Teams define semantic models and metrics in YAML, keeping them version-controlled and connected to lineage. MetricFlow turns those configurations into SQL at query time and handles joins, so downstream tools can use common definitions rather than each maintaining its own metric logic. That is a strong fit when consistency across teams and products matters; it is less compelling if a team does not want its metrics tied to dbt.

MetricFlow also rewrites queries to reduce compute and latency, optimizes joins, and caches repeated queries. These are useful workload optimizations, although the product does not promise a particular performance outcome. GraphQL and JDBC APIs and a Python SDK provide ways to query metrics and dimensions. An API, the dbt MCP Server and out-of-the-box integrations cover BI, notebooks, spreadsheets, embedded analytics and AI systems. Claude and ChatGPT can connect to governed metrics through MCP.

Supported data platforms are Snowflake, BigQuery, Databricks and Redshift. The service is cloud-deployed and supports web and API access. Security and compliance badges include SOC 2 Type II, ISO 27001:2022, ISO 27017, ISO 27018, ISO 42001, GDPR and CCPA. Starter includes 24x5 support; Enterprise offers priority support with SLAs.

Pricing

The Developer plan is free and includes one developer seat, one project and 3,000 successful models per month, but no Semantic Layer. It is useful for developing with dbt, not for adopting this product.

Starter costs $100 a month, billed per user/month. It includes five developer seats, one project, 15,000 successful models per month and 5,000 queried metrics per month, as well as Semantic Layer access. This is the entry point for teams that want the product, but the per-user price and metric quota make it important to estimate both team size and query demand.

Enterprise and Enterprise+ use custom pricing. Both include custom developer seat counts, 100,000 successful models per month and 20,000 queried metrics per month. Enterprise includes 30 projects; Enterprise+ allows unlimited projects. These tiers suit organizations needing more model capacity, project scale or enterprise support, but their custom pricing makes a direct cost comparison difficult.

Accessing or defining Semantic Layer metrics requires Starter or Enterprise-tier access. Single-tenant accounts also need setup and enablement from an account representative. The free plan therefore does not provide a no-cost way to query Semantic Layer metrics.

Platforms

dbt Semantic Layer runs in the cloud and is available through web and API platforms. Its supported warehouse platforms are Snowflake, BigQuery, Databricks and Redshift.

Who it's for

This is a practical choice for dbt-based data teams that need governed metrics reused across BI, embedded analytics, notebooks, spreadsheets or AI workflows. Teams should look elsewhere if they need Semantic Layer querying on a free plan, do not use a supported warehouse, or expect to exceed Starter's 5,000 queried metrics per month without moving to a custom-priced tier.

Pros and cons

• Pro: YAML definitions, version control and lineage give teams a consistent way to manage shared metric logic.
• Pro: GraphQL, JDBC, a Python SDK and broad integration targets make metrics accessible to both analytics tools and applications.
• Pro: MCP access connects Claude and ChatGPT to governed metrics rather than leaving AI tools to recreate definitions.
• Con: The free Developer plan excludes Semantic Layer, so using the product starts at $100 a month per user.
• Con: Starter's 5,000 queried metrics per month may constrain wider use, while higher quotas require custom-priced plans.
• Con: Its dbt account requirement and four supported warehouse platforms limit its fit for teams outside that ecosystem.

Alternatives

Cube is worth considering for teams that want a free forever entry point with up to five workbooks, 1,000 daily requests and one-day query history retention, or need Linux and self-hosted platforms.

Starburst Enterprise Context Layer offers a free plan for up to three clusters with standard cluster execution for ad hoc queries, making it an option for teams whose starting need is that cluster-based query setup.

Definite may suit a small team seeking a free plan with two users, two connectors, 1 GB of storage, five monthly credits and daily sync.

Sema is another freemium option with a free forever Sandbox plan.

Strata offers a free plan with one developer, 25 users, two data sources per project and unlimited projects and branches, a different fit for teams prioritizing those limits.

Verdonz has a free guided workspace with sample data, but its free plan does not allow teams to use their own business data.

MetricFlow is the free, Apache 2.0-licensed option for teams prepared to use it with a working dbt project and dbt adapter.

Malloy is a free alternative.

Browse more options in Semantic Layer Software, Database Documentation Tools and Data Transformation Tools.

Verdict

Choose dbt Semantic Layer if your team already relies on dbt and needs version-controlled, governed metrics available consistently across analytics, applications and AI. Shared definitions and multiple query interfaces are its strongest reasons to buy; the paid access requirement, Starter quota and dbt-and-warehouse fit are the reasons to look elsewhere.