Databricks connector

Query Your Databricks Data with AI

Connect your Databricks SQL warehouse to Better Analyst and ask questions about your Unity Catalog tables in plain English. No SQL required.

Free to tryRead-only accessEncrypted credentials

Try a real question

Quick setup

Enter your workspace hostname, SQL warehouse HTTP path, and an access token or service principal. Connect and start querying in under a minute.

Secure connection

Encrypted credentials with read only as the default. Every query is checked as a single read-only SELECT, and optional destination writes are restricted to an explicitly approved schema.

Real-time access

Query your live Delta tables and views directly through your SQL warehouse. No exports or data movement needed.

What You Can Analyze

Lakehouse Queries

  • “What schemas and tables are in this catalog?”
  • “Show me a sample from the orders table”
  • “Total revenue by region last quarter?”

Cross-Schema Joins

  • “Join marketing events with sales conversions”
  • “Which campaigns drove the most revenue?”
  • “Customer lifetime value by acquisition source?”

Dashboards & Alerts

  • “Build a weekly KPI dashboard from the gold tables”
  • “Alert me when daily orders drop below normal”
  • “Refresh this chart every morning”

Schema Exploration

  • “What tables are in the analytics schema?”
  • “Show column types for the orders table”
  • “Which tables have the most rows?”

Smarter Over Time

Teach the AI your data

Upload documentation and provide feedback to make every query more accurate for your Databricks data.

Knowledge Base

Upload schema definitions, data dictionaries, and business logic docs to help the AI understand your Databricks structure.

Uploaded files

schema.sql

24 KB

data-dictionary.pdf

156 KB

business-logic.md

8 KB

SQL Feedback Loop

Rate generated queries as correct or incorrect. Your feedback is saved and used to improve future Databricks queries.

Generated SQL

SELECT product, SUM(revenue)
FROM sales
GROUP BY product
ORDER BY 2 DESC

Was this correct?

How it works

1

Upload context

Add schema docs, data dictionaries, or SQL files that describe your Databricks tables and business logic.

2

Ask questions

Query your data in plain English. The AI uses your uploaded context to write more accurate SQL.

3

Refine with feedback

Mark queries as correct or incorrect. The AI learns from your feedback and improves over time.

Explore Other Connectors

Google Analytics

Connect Google Analytics 4 to Better Analyst and analyze website traffic, user behavior, and conversions using AI-powered natural language queries.

Google Search Console

Connect Google Search Console to Better Analyst and analyze search performance, keyword rankings, and indexing with AI-powered queries.

Google Ads

Connect direct and MCC-managed Google Ads accounts to analyze campaigns, keywords, search terms, costs, and conversions with AI.

Meta AdsComing soon

Coming soon: analyze Facebook and Instagram campaign spend, reach, clicks, conversions, and ROAS with AI.

Tableau

Connect Tableau Cloud to Better Analyst and chat with synced dashboards, views, and datasources as BI collections.

Power BI

Connect Power BI to Better Analyst and analyze semantic models with DAX while browsing synced workspaces, reports, and dashboards.

Airtable

Connect a read-only Airtable base and analyze selected tables in chat and refreshable dashboards.

HubSpot

Connect a HubSpot portal read-only and analyze contacts, companies, deals, and tickets in chat, refreshable dashboards, alerts, and Data Flows.

Google Sheets

Connect Google Sheets to Better Analyst and sync spreadsheet data for AI-powered analysis, dashboards, reports, and scheduled refreshes.

Google Drive

Connect Google Drive to Better Analyst and import spreadsheets, CSVs, and data files for AI-powered analysis without manual downloads.

Dropbox

Connect Dropbox to Better Analyst and import Excel, CSV, and TSV files for AI-powered analysis without manual downloads.

OneDrive

Connect Microsoft OneDrive to Better Analyst and import Excel, CSV, and TSV files directly into AI chats, dashboards, and reports.

SharePoint

Connect Microsoft SharePoint to Better Analyst and import Excel, CSV, and TSV files from document libraries for AI-powered analysis.

Google BigQuery

Connect Google BigQuery to Better Analyst and analyze massive datasets in your data warehouse using AI-powered natural language queries.

Snowflake

Connect your Snowflake data warehouse to Better Analyst and analyze your data with AI-powered natural language queries. No SQL required.

MySQL

Connect your MySQL database to Better Analyst and query your data using AI-powered natural language. No SQL expertise required.

PostgreSQL

Connect your PostgreSQL database to Better Analyst and explore your data with AI-powered natural language queries. No SQL required.

Supabase

Connect your Supabase Postgres database to Better Analyst and analyze live product, user, and event data with AI-powered natural language queries.

Microsoft SQL Server

Connect Microsoft SQL Server to Better Analyst and explore enterprise data with AI-powered natural language queries. No SQL expertise needed.

Amazon Redshift

Connect your Amazon Redshift data warehouse to Better Analyst and analyze your data with AI-powered natural language queries. No SQL required.

Salesforce

Connect a Salesforce org read-only and analyze accounts, opportunities, leads, cases, and custom objects in chat, dashboards, alerts, and Data Flows.

Frequently Asked Questions

Better Analyst connects to Databricks SQL warehouses (serverless, pro, or classic) on AWS, Azure, and Google Cloud, and reads Unity Catalog catalogs, schemas, tables, and views.
Your workspace hostname, the HTTP path of a SQL warehouse, and either a personal access token or an OAuth service principal (client ID and secret). We recommend a dedicated service principal with only USE CATALOG, USE SCHEMA, and SELECT on the data you want to analyze.
In Databricks, open SQL Warehouses, choose your warehouse, and open the Connection details tab. Copy the Server hostname and HTTP path from there.
Yes. Credentials are encrypted at rest and all connections use TLS. Read only is the default: every query is validated as a single read-only SELECT, and recurring Data Flows additionally verify that the principal holds no write or ownership privileges.
Queries run on the SQL warehouse you choose and use its compute like any other query. A small serverless warehouse with auto-stop keeps costs low and isolates Better Analyst from production workloads.

Ready to Query Your Databricks Data with AI?

Connect Databricks