Technical Guide: Generating Business Intelligence with Basedash

Practical Summary and Prerequisites

Basedash is an AI-native analytics platform designed to provide self-serve business intelligence by connecting to over 750 data sources. It functions as a semantic layer and managed warehouse that allows users to query data using natural language, generate dashboards, and automate insights. The platform is built to ensure auditability by providing the SQL behind every AI-generated answer, ensuring that business intelligence remains governed and traceable.

To utilize Basedash effectively, users must have access to their data sources, which can include databases like Snowflake, BigQuery, or Databricks, as well as SaaS platforms like Stripe, Salesforce, or HubSpot. While no complex migration is required, users should be prepared to define metrics and business logic within the platform's semantic layer to ensure the AI provides accurate, context-aware responses. The platform supports SSO and SCIM for enterprise-grade security and governance.

Connecting Data and Preparing the Semantic Layer

The foundation of Basedash is its ability to ingest data from diverse sources. Users can connect existing warehouses or utilize the platform's managed DuckDB warehouse, which keeps tables synced and structured for agent-based querying. Once connected, the platform allows for the definition of metrics, segments, and business logic. This semantic layer is critical because it acts as a self-improving context engine; the AI learns from previous conversations and defined skills, ensuring that subsequent queries are grounded in the company's specific business definitions.

For example, defining a metric like "MRR" involves specifying the logic for active subscription revenue, excluding internal accounts or trials. By centralizing these definitions, Basedash ensures that every team member, regardless of technical skill, queries the same governed data. This setup process is designed to be completed in under 30 minutes, allowing teams to move quickly from raw data connection to actionable insights without building complex ETL pipelines.

Documented Workflow for AI-Driven Analytics

The core workflow involves interacting with the AI analyst through natural language. Users can ask questions such as "Build me a dashboard for revenue and signups over the last year" or "What was net revenue retention by segment last quarter?" The platform processes these requests by scanning the data catalog, identifying relevant tables, and generating the necessary SQL. Every result is accompanied by the underlying SQL and the sources used, providing full transparency.

Beyond ad-hoc queries, Basedash supports proactive analytics through real-time observability. Agents monitor metrics around the clock and can trigger alerts or post briefings to Slack or email when anomalies are detected. This allows teams to monitor their business performance with the same rigor applied to production systems. The platform also supports embedding these capabilities directly into internal products using React components, as shown in the following example:

<BasedashProvider fetchToken ={getToken} > <BasedashDashboards /> <BasedashChat /> </BasedashProvider>

Interoperability and Advanced Integration

Basedash is designed to be fully interoperable, allowing teams to use the platform's native interface or integrate it with external AI clients. Through the MCP (Model Context Protocol) server, users can connect tools like Claude, ChatGPT, or Cursor to their governed data. This ensures that even when using external agents, the data remains secure and governed by the permissions set within Basedash. The platform also provides a REST API for programmatic access to its chat and analytics features.

For developers looking to integrate Basedash into their own workflows, the MCP server can be added to client configurations. For instance, adding the server to a Cursor environment allows for data-aware coding and analysis directly within the editor. This flexibility ensures that organizations are not locked into a single interface while maintaining the integrity and security of their underlying data stack.

Limitations and Governance Considerations

While Basedash provides powerful automation, it is subject to the quality of the underlying data and the accuracy of the defined semantic layer. Users must ensure that metrics are correctly governed to avoid misleading AI outputs. Security is a primary focus, with features including role-based access control, row-level security, and audit logs that track every query and configuration change. The platform is SOC 2 Type II compliant, and customer data is explicitly stated not to be used for training models.

Organizations should use Basedash when they require a unified view of disparate data sources and want to empower non-technical stakeholders to perform self-serve analytics. It is particularly well-suited for teams that need to move away from manual report generation and toward real-time, AI-driven business intelligence. For more information on the platform's capabilities, visit https://www.basedash.com.