Purpose
Basedash is positioned as a "data analytics platform for the AI era," and its official description says it includes self-service BI, embedded analytics, ETL, a managed data warehouse, and a semantic layer to meet AI analytics needs. It is aimed at scenarios where teams, customers, or agents want to query data directly, and the official claim is that 5,000+ teams already use it.
Main capabilities
- Self-service analytics: Team members can ask questions in natural language and get charts, dashboards, or direct answers; results are based on governed definitions and show the underlying SQL for auditing and traceability.
- Proactive analytics: Agents continuously monitor metrics, flag changes and their causes, and can send briefings to Slack or email; supports scheduled reports or triggering when data changes.
- Embedded analytics: Embed dashboards and AI conversations into a product with a small amount of React code, white-label them as your own brand, and scope each customer to their own data rows via JWT SSO.
- Semantic layer and context engine: Define metrics, cohorts, and business logic as reusable models, or connect an existing dbt project; after adding skills and team context, the AI retains what it learns in conversations.
- Managed data warehouse: After connecting data sources, Basedash provides a managed DuckDB warehouse that keeps tables synchronized, structured, and queryable by agents, with storage scaling and point-in-time recovery; you can also use an existing Snowflake, BigQuery, or Databricks instead.
- Data source connections: The official claim is support for 750+ data sources, including Salesforce, HubSpot, Stripe, Shopify, and others, keeping data fresh after a single authentication; you can also connect your own databases and data warehouses directly.
- Interoperability: Connect Claude, ChatGPT, or Cursor via MCP, call the REST API from your own app, or ask questions in Slack; the official claim is that all paths use the same models and permissions.
- Security and governance: AI usage is centralized in a governed environment, questions run under user permissions and are logged for audit; it provides AI usage visibility and optional spend caps, audit log streaming to SIEM, source tables and SQL shown for every answer, approvals for write queries, SAML 2.0/OIDC SSO and SCIM, role permissions plus row-level security, SOC 2 Type II audit, and self-hosting within a user VPC (air-gap capable and using your own model keys).
How to use
Official pages do not provide specific installation commands or detailed operating steps. Available information includes: the page mentions "create an account" or "book a demo" to explore data; embedded analytics requires "a few lines of React"; it can be accessed via MCP, REST API, or Slack; the official claim is that you can "get started in 30 minutes," and it says it can help migrate data and dashboards from other tools.
Use cases
- Teams querying data internally in natural language and generating charts or dashboards.
- Embedding analytics capabilities and AI conversations into your own product and offering them to customers.
- Aggregating answers across databases, PostHog, Stripe, and other sources.
- Continuous monitoring and anomaly detection for metrics, with pushes to Slack or email.
- Data governance scenarios requiring auditing, permissions, row-level security, and self-hosting.
Limitations
Official materials do not state specific pricing, free quotas, the complete list of supported data sources, model vendor details, or SLA. The specific configuration steps and requirements for some features (such as MCP, self-hosting, and air-gapping) are not yet provided.