This n8n workflow automates a smart shopping assistant that leverages OpenAI’s language models, Retrieval-Augmented Generation (RAG), and WooCommerce to enhance customer experience in an online shoe and accessories store. When a chat message is received via webhook, the workflow processes the conversation to determine if the customer is searching for a product or asking general questions.
The process begins with the ‘When chat message received’ trigger, capturing user input. The ‘Information Extractor’ analyzes the message to identify product-related intent, extracting crucial details such as keywords, price ranges, SKUs, and category preferences. This information is passed to the ‘personal_shopper’ WooCommerce node, which fetches relevant products based on the user’s criteria.
Simultaneously, a more general inquiry is handled by a RAG system, which utilizes documents stored in Google Drive, embedded and indexed via Qdrant vector database. Texts are split into manageable chunks, embedded with OpenAI, and stored in Qdrant for efficient retrieval. Depending on the user’s query, the AI agent dynamically decides whether to activate the product filtering or address non-product questions using the RAG system.
Additional nodes handle memory management, document handling, embedding, and communication with external services like Google Drive, Qdrant, and WooCommerce, making this workflow a comprehensive solution for a conversational, AI-driven shopping assistant that improves online customer support and product discovery.
This workflow is particularly useful for eCommerce sites aiming to provide personalized shopping experiences through intelligent chatbots that dynamically recommend products or answer store-related questions.
Reviews
There are no reviews yet.