Query Business Data Chatbot with RAG & Text-to-SQL via Peliqan

This n8n workflow powers an intelligent chatbot that interacts with business data from diverse sources such as Notion, Chargebee, HubSpot, and more. It leverages Peliqan.io as a data cache with one-click ELT to sync data into a warehouse, enabling fast RAG searches via Supabase vector store and precise Text-to-SQL queries. The AI Agent intelligently decides whether to use RAG for unstructured knowledge questions (e.g., Notion pages), Text-to-SQL for analytical queries (e.g., customer revenue and support tickets), or a hybrid approach.

Key benefits include seamless integration of multiple data sources without complex ETL setups, reduced query latency through Peliqan caching, and accurate responses powered by OpenAI. It saves significant time for business users by automating data retrieval and analysis in natural language, eliminating manual SQL writing or document searches.

Ideal use cases: Customer support teams querying CRM data, sales analyzing revenue trends, or executives getting instant insights from docs and databases. Prerequisites are a Peliqan free trial account and connected data sources; setup involves adding sources in Peliqan and configuring n8n credentials for Supabase and OpenAI.