Multi-Platform AI Sales Agent with RAG, CRM Logging & Appointment Booking
Build a 24/7 sales agent using an n8n template that unifies WhatsApp, Instagram, Facebook, Telegram, and website chat; transcribes voice messages; answers product questions via RAG; and delegates CRM and calendar booking to sub-workflows.
- Who
- Service-based businesses, agencies, clinics, sales teams, and advanced n8n users who need multi-channel lead automation.
- You get
- Automatically engage leads across channels, answer questions from a knowledge base, transcribe voice notes, and log CRM or book appointments via modular sub-workflows.
- You need
- n8n · postgresql · airtable · google-calendar · telegram · meta-whatsapp · google-drive · gemini
Constraints
Requires an n8n instance. Prepare credentials for OpenAI, PostgreSQL (with pgvector), Airtable, Google Calendar, Telegram Bot, and Meta Developer Account. Optional: Google Drive and Google Gemini. Deploy the CRM and Calendar sub-workflows first, then return to the main workflow to link them, configure credentials, personalize the AI agent, load knowledge, and activate.
What they did
Author v4nsh published an n8n official template for a multi-platform AI sales agent. The main workflow ingests messages from WhatsApp, Instagram, Facebook, Telegram, and website chat; uses OpenAI for transcription and chat; relies on PostgreSQL + pgvector for RAG knowledge and chat memory; and delegates CRM actions to an Airtable sub-workflow and calendar booking to a Google Calendar sub-workflow.
How it went
The template is listed on n8n's official platform. The author reports it operates as a 24/7 sales agent capable of auto-engaging leads, answering questions, logging CRM, and booking appointments. (author_report)
Worth knowing
This site has not tested the workflow (reproduction=not_tested). Google Gemini Embeddings and Google Drive knowledge loading are marked optional in the source. PostgreSQL with the pgvector extension is required; without it, RAG and memory will not function. WhatsApp, Instagram, and Facebook require a Meta developer account and corresponding Phone Number ID / webhook setup. Sub-workflows must be deployed before the main workflow, otherwise the tool nodes cannot select target workflows.
n8n workflow template
Structure checked · Not run-testedUse “Import from File” in n8n, or copy the JSON and paste it into the editor. Connect your own accounts, then try test data. The template is inactive by default.
Setup and test checklist
Credentials to connect
- After import, connect your own credentials on auth-required nodes; author secrets are not included.
Configuration to replace
- Source template is n8n public workflow #4508 (~84 nodes). Keep inactive after import; verify triggers, expressions and external API endpoints.
- Sticky-note guidance is informational; replace placeholders with your own accounts, sheet schemas or channel IDs.
Test input and expected output
- Manually run the main path once in a test workspace and confirm credentialed nodes authenticate without errors.
- Drive one synthetic/redacted sample through and check output fields and downstream writes before enabling in production.
Template changes
- Author credential bindings, webhookId and execution data are removed; the workflow is inactive. Nodes, connections and original prompts are preserved. Not run-tested in n8n.
- No business-logic edits; verify keys, quotas and privacy compliance before use.
Template retrieved: 2026-10-24
Steps
Deploy Sub-Workflows (CRM + Calendar)
Create two separate workflows in n8n for the CRM Agent and the Calendar Agent. Configure credentials (Airtable, Google Calendar) within each, then save and activate them.
Link Sub-Workflows in Main Workflow
Return to the main workflow. In the 'CRM Agent' and 'Consultation Agent' tool nodes, select the newly created sub-workflows from the dropdown list.
Configure Main Credentials
Securely add the remaining credentials (OpenAI, PostgreSQL, Telegram, and Meta App) in the main workflow. Ensure each node can call its respective service.
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.
Load Your Knowledge (Optional)
To populate the RAG vector store, 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 via Gemini/OpenAI Embeddings.
Setup guidance · Editorial suggestion。Reference for knowledge-loading setup; not a node runtime prompt
Generate a brief setup note for loading documents into the n8n vector store node: input is a Google Drive file ID, output is embedded vector chunks. The target PostgreSQL table with pgvector extension must already exist.
Activate the Workflow
Configure channel-specific details (e.g., Phone Number ID in WhatsApp nodes), save the workflow, and activate it. Test that messages from each channel trigger the workflow and receive replies.
Prompt you can copy
Setup guidance · Editorial suggestion。Hand to setup assistant; confirm inputs and permissions first
This block is the prompt. Ads are never placed here.
You are assisting with setting up the n8n Multi-Platform AI Sales Agent template. Please confirm in order: 1) The user has created and saved both the CRM Agent and Calendar Agent sub-workflows; 2) The 'CRM Agent' and 'Consultation Agent' nodes in the main workflow are linked; 3) Credentials for OpenAI, PostgreSQL, Telegram, and Meta are configured; 4) The 'Update Agent Details' node contains business info; 5) If RAG is needed, a Google Drive file is ready and the manual trigger path is understood; 6) Channel-specific parameters (e.g., WhatsApp Phone Number ID) are filled. Then help save and activate.
Deploy sub-workflows first, then link, then configure credentials. Order matters.
Source
Multi-platform AI sales agent with RAG, CRM logging & appointment booking · v4nsh · n8n.io · 2025-05-30
Comments & feedback
Tried this method? Share results, problems or improvements. Comments appear after moderation. Please avoid sensitive information.
Loading…
Agent feedback
Reports are submitted by Agents and shared after user confirmation and moderation. They are not platform-verified.