💬 Multi-platform AI sales agent with RAG, CRM logging & appointment booking

⚡ 5,259 views · 💬 Lead Nurturing & AI Sales Agents

💡 Pro Tip — If you’ve run into Facebook’s Graph API permission restrictions, ScraperNode is a community node that scrapes pages, groups, posts, and comments directly — no app review needed.

Description

Multi-Platform AI Sales Agent with RAG, CRM Logging & Appointment Booking

This workflow acts as a 24/7 sales agent, engaging leads across WhatsApp, Instagram, Facebook, Telegram, and your website. It intelligently transcribes audio messages, answers questions using a knowledge base, and uses modular sub-workflows to manage your CRM in Airtable and book appointments in Google Calendar.

Who Is It For?

This solution is designed for service-based businesses that require a sophisticated, multi-channel lead management and appointment booking system.

• Agencies, consultants, and clinics looking to automate lead qualification and booking across text and voice messages.
• Sales teams needing a robust system that separates CRM and calendar logic for easier management and scalability.
• Advanced n8n users who want a powerful, agent-based framework that can be easily customized and extended.

Key Features & Workflow

This system is composed of a parent workflow that orchestrates the conversation and calls specialized sub-workflows (agents) to perform specific tasks.

• Omni-Channel Engagement: Greets and converses with leads naturally on Website Chat, WhatsApp, Instagram, Facebook, and Telegram.
• Voice Message Transcription: Automatically transcribes incoming audio/voice notes from WhatsApp and Telegram into text, allowing the AI to understand and respond without manual intervention.
• RAG-Powered Intelligence: The AI agent draws answers from your dedicated PostgreSQL Knowledge Base (RAG), providing accurate information about your services, pricing, and process. A built-in tool allows you to load new knowledge documents from Google Drive.
• Modular CRM Agent (Sub-Workflow): When the conversation requires CRM actions (like creating a contact or updating an opportunity), the main agent calls a dedicated Airtable CRM sub-workflow to handle the task.
• Modular Calendar Agent (Sub-Workflow): To book appointments, the main agent calls a separate Google Calendar sub-workflow, which manages event creation, updates, and deletions.

What You’ll Need

This workflow framework requires credentials for the following services:

• An AI provider (OpenAI is used for the main agent, transcription, and embeddings).
• PostgreSQL database with the pgvector extension (for RAG and chat memory).
• Airtable account (for the CRM sub-workflow).
• Google Calendar account (for the Calendar sub-workflow).
• Telegram Bot credentials.
• A Meta Developer Account (for WhatsApp, Instagram & Facebook).
• (Optional) Google Drive for loading knowledge documents.
• (Optional) Google Gemini for embeddings.

How to Set It Up

This is a modular system. You must set up the sub-workflows first.

1. Deploy Sub-Workflows: Create two new, separate workflows using the provided templates for the CRM Agent and the Calendar Agent. Configure the credentials (Airtable, Google Calendar) within those workflows, then save and activate them.
2. Link Sub-Workflows in Main Workflow: Return to this main workflow. In the “CRM Agent” and “Consultation Agent” tool nodes, select the new workflows you just created from the dropdown list.
3. Configure Main Credentials: Securely add the remaining credentials (OpenAI, PostgreSQL, Telegram, and your Meta App) in this main workflow.
4. Personalize the AI Agent: In the “Update Agent Details” node, fill in your business name, the agent’s persona name, and your key services to customize the AI’s personality.
5. Load Your Knowledge (Optional): To populate the RAG database, use the manual trigger path starting with “When clicking ‘Execute workflow’” to download a document from Google Drive and insert it into your PostgreSQL vector store.
6. Activate: Configure your channel-specific details (e.g., Phone Number ID in WhatsApp nodes), save the workflow, and activate it.

Demo CRM Table: View on Airtable

🔗 Nodes Used

Airtable, HTTP Request, Webhook, Telegram, Telegram Trigger, Google Drive

📥 Import

Download workflow.json and import into n8n: Workflow menu → Import from File

📖 Importing guide · 🔑 Credential setup