Unlocking analyst-driven data transformation: dbt Canvas is GA

Today, dbt Canvas, an all-new, AI-powered visual editing experience in dbt is generally available.

With dbt Canvas, data analysts can finally take an active role in data transformation. They can use an intuitive drag-and-drop interface that’s seamlessly integrated into the governed, scalable analytics development lifecycle and powered by context-aware AI. It empowers teams to move faster, reduce bottlenecks, and collaborate more effectively without compromising on quality, trust, or version control.

You can get started with dbt Canvas today—just open a project and start building.

Making analytics a team sport with dbt Canvas

At dbt Labs, we believe data is a team sport.

For nearly a decade, dbt has helped organizations bring software engineering best practices like version control, testing, and CI/CD to analytics engineering. But that was just the beginning. The full promise of data can only be realized when all data practitioners, not just data and analytics engineers, can contribute to their organization’s data estate in an accessible, yet governed way. dbt Canvas, now GA, makes this promise a reality.

Self-service analytics requires more than just access to data; it requires governance

For years, data leaders have chased the promise of self-service analytics. But true self-service doesn’t mean simply handing analysts access to datasets or procuring visualization tools. It means enabling them to contribute to a mature analytics workflow—the ADLC—where data is shaped, defined, and prepared for the entire organization.

Data analysts are closer to the business questions than anyone else on the team. They know the “why” behind the metrics. They understand the quirks in the data. They have a deep understanding of evolving business strategies and goals. But when it comes to contributing directly to data transformation, they’re often blocked.

Why?

Because data teams are overwhelmed. Engineers are buried in tickets and ad-hoc requests. Analysts wait in queues. To move faster, they turn to ungoverned low-code editors, SQL editors, BI tools, or scripting in isolation, creating shadow pipelines.

We’ve heard the frustration loud and clear: Data teams need a way for analysts to safely, confidently, and effectively contribute to the data transformation process without sacrificing governance or quality. And in the AI era, analysts need trustworthy AI that's powered by the full context of their data assets to contribute faster.

Empower analysts to build with confidence and control

dbt Canvas is an AI-powered visual experience built for analysts. It enables data practitioners across a spectrum of technical skills to build and collaborate on data transformations without needing to write SQL from scratch.

Build and edit dbt models visually - no SQL required

With dbt Canvas, analysts start by exploring the input data that is available, whether it’s curated models or sources imported from the warehouse, or an uploaded CSV file. They can drag and drop these data sets directly onto the canvas, where they can preview the data and associated output columns. Analysts may not know the data that’s available to them. Canvas includes robust search and discovery to find existing datasets and always-on data profiling capabilities that help you understand that data.

From there, analysts can apply transformations. Common operations like joins, aggregations, filters, calculated columns (formulas), sorts, and unions are all available as modular building blocks. Each transformation is visually represented as a node in the canvas, making it simple to trace logic, understand relationships, and collaborate with teammates.

As analysts build out a model, they can preview the data output at every stage, seeing exactly how each transformation impacts the dataset. This step-by-step feedback loop builds confidence, reduces errors, and accelerates iteration.

What makes dbt Canvas truly powerful is what’s happening under the hood. Every transformation, filter, and calculation built in Canvas is automatically compiled into SQL that adheres to your organization’s dbt project structure and coding standards.

This means the visual logic analysts create isn’t locked into a proprietary format, it becomes production-grade dbt code that’s version-controlled, testable, and reviewable just like any other model in your dbt project.

Built-in governance with Git and PR workflows

Every model built in Canvas can be committed to your dbt project using the same version control workflows you already use in dbt Studio or the VS Code extension.

Even if you’re not a git expert, Canvas makes it simple: click to commit, and submit a pull request for review. Analysts become first-class contributors to the analytics engineering workflow - with all the safeguards teams require.

Accelerate development with dbt Copilot

Canvas includes dbt Copilot, our built-in, context-aware AI assistant, which can help generate SQL expressions in formula nodes, explain specific steps in the workflow, modify existing models, or create brand-new models from scratch—just by describing what you want in natural language.

Copilot captures project context, like model relationships, tested logic, and naming conventions, from your dbt project to power Canvas, helping analysts build more accurate, trusted models using natural language and visual workflows

Whether you’re exploring a data source or defining a new KPI, Copilot helps you move from idea to model faster than ever.

Whether you're a data analyst building your first model or a senior analytics engineer reviewing a pull request, dbt Canvas ensures that what’s created visually can be integrated directly into your governed analytics workflow. And because everything in Canvas aligns with dbt’s underlying principles—modularity, reusability, and transparency—teams get the best of both worlds: accessibility for analysts, and accountability for engineering.

Coming soon to dbt Canvas

We’re just getting started. Here’s what’s next:

• Support for additional SQL operations, including window functions and advanced filter expressions
• Connect and import data to the warehouse from other sources, like Google Sheets.
• Define dbt tests and unit tests directly in Canvas

Making dbt the standard for analysts

dbt Canvas is part of a growing set of new AI-powered self-service capabilities we’re launching to empower analysts to participate in dbt's governed workflows with confidence and clarity. New capabilities include:

• dbt Insights (now in Preview) helps analysts ask any ad-hoc question and get an answer fast from governed data, all within dbt. Using SQL or natural language prompts, analysts can create queries, validate models, generate visualizations, and easily share findings, all without writing complex code. Transparent governance and full lineage ensure trust and deliver high‑quality insights in one seamless workflow. With Insights, you can:
  • Write, run, validate, and iterate on SQL queries without switching tools using features like dbt Copilot, syntax highlighting, and query history.
  • Leverage dbt metadata, trust signals, and model lineage to write better, faster queries and quickly identify issues and root causes.
  • Make data accessible to a wide range of users, from SQL pros to AI-assisted analysts, using both code and context-aware AI.
  • Go from question to model faster by validating ideas before formalizing them in dbt Canvas or the dbt Studio (IDE).
  • Quickly explore patterns, trends, or anomalies in your data with built-in visualizations, no need to export to external tools
• New functionality in dbt Catalog (formerly known as dbt Explorer) allows users to go beyond their dbt estate, and search and explore Snowflake assets—like tables and views—directly within dbt. Now, analysts and developers can build holistic context for their data estate without switching platforms or tabs.

Together, these new platform experiences reflect our belief that the future of analytics engineering is collaborative, governed, and accessible to every data practitioner. Not only are we introducing new capabilities, but we’re also making dbt Cloud more accessible than ever with our new flexible pricing. This approach is designed to make it easy to expand access to your entire analyst team.

Get started today

Check out the below resources to get your teams started with Canvas:

• Explore the docs
• Take the dbt Canvas Fundamentals training course to ramp up fast
• Learn more about dbt for analysts
• Join our webinar in late June for an in-depth look at the new features tailor-made for analysts