What is BigQuery data canvas?
BigQuery data canvas is a Gemini-powered visual workspace inside BigQuery Studio. Instead of writing SQL by hand, users express their intent in natural language, for example “Show me the top 10 customers by revenue”. BigQuery data canvas generates the matching SQL query, runs it and renders the result. The analysis takes shape on a canvas as a directed acyclic graph (DAG), where each node represents a step: search, table, SQL, visualization or insight.
BigQuery data canvas solves the problem that exploratory data analysis is often slow and hard to follow. The visual, node-based approach makes every analysis step visible and reproducible, while the natural-language interface speeds up the entry point. Data teams can test hypotheses faster, document queries and visualizations as one connected workflow, and reuse results directly from within BigQuery.
Core Features
- Natural-language search and NL-to-SQL: Find tables, views and materialized views through the data catalog using natural language, and generate and run queries from text prompts.
- Node-based DAG workflow: Build analyses from typed nodes (text, search, table, SQL, destination, visualization, insights), so dependencies and steps stay visible as a graph.
- Visualization and automated insights: Create charts such as bar, line, pie, scatter and heat map from a prompt, plus automatically generated data insights to surface patterns and anomalies.
- Gemini chat and export: A Gemini chat assistant creates and runs nodes; canvas version control plus export to notebooks, scheduled queries or PNG charts.
Typical Use Cases
Exploratory data analysis: Analysts test hypotheses through natural language without writing SQL by hand for every step, while keeping the generated queries visible for review.
Prototyping queries and reports: Queries and visualizations come together quickly on the canvas and can then be reused as a scheduled query or notebook.
Data discovery in the catalog: Through natural-language search, teams find relevant tables and views in the catalog even when they do not know the exact names.
Benefits
- Faster entry into data analysis through natural language instead of manual SQL
- Traceable, reproducible analysis workflows through the visual DAG approach
- No separate license: integrated into BigQuery Studio, billed through Gemini in BigQuery and BigQuery compute
- EU regions available, with Gemini processing kept in the EU for EU datasets
Integration with innFactory
As a certified Google Cloud Partner, innFactory supports you with the adoption and operation of this service.
Typical Use Cases
Frequently Asked Questions
What is BigQuery data canvas?
BigQuery data canvas is a Gemini-powered feature inside BigQuery Studio. It lets users find, transform, query and visualize data using natural language. The analysis is built on a visual canvas as a directed acyclic graph (DAG) of nodes, so each step stays visible and traceable.
When should I use BigQuery data canvas?
BigQuery data canvas suits exploratory data analysis, rapid prototyping of queries and visualizations, and finding the right tables through the data catalog. It targets data analysts and data engineers with basic SQL knowledge who want to reach results faster and capture them as a documented workflow.
How much does BigQuery data canvas cost?
BigQuery data canvas is part of Gemini in BigQuery. The AI features are billed either through pay-as-you-go AI processing fees or a Gemini Code Assist subscription. On top of that, standard BigQuery costs apply for storage and compute (query processing). Current pricing is listed on the official BigQuery pricing page.
What are the requirements and limitations?
An administrator must enable Gemini in BigQuery at the project level, and users need the BigQuery Studio User and Gemini for Google Cloud User IAM roles. Limitations apply to BigQuery ML, Spark, nested and repeated fields, JSON and complex data types; geomap charts are not supported. Basic SQL familiarity is recommended.
