Multi-Channel AI Customer Support Agent with Gmail and Telegram - n8n Workflow

• Small to medium businesses needing 24/7 support.


• E-commerce stores looking to reduce response times.


• SaaS companies automating FAQ handling.


• Teams seeking scalable, consistent, multi-channel customer service.


• Users looking for advanced n8n templates integrating Langchain features.

Customer support demands speed and consistency. This sophisticated n8n workflow dramatically improves efficiency by creating two parallel, channel-specific AI agents for Gmail and Telegram. Each path uses an n8n node configured to leverage conversational memory and a defined knowledge base stored in Google Sheets. The primary value of this n8n workflow is its ability to contextually understand customer needs, provide immediate and accurate responses, and ensure complex or escalated inquiries are logged instantly for human review. By utilizing these specialized n8n templates, businesses can handle routine questions automatically, freeing up human agents for critical issues. Every step, from the initial n8n trigger to the final logging, is designed for maximum performance and reliability.

This automation begins with one of two distinct n8n trigger nodes, launching parallel execution paths:

1. Gmail Path Trigger: The Gmail Trigger n8n node monitors for unread emails in the designated inbox.


2. Telegram Path Trigger: The Telegram Trigger n8n node activates instantly upon receiving a new message.

Once triggered, the workflow standardizes the input data using a Set n8n node, extracting critical information like customer name, ID, and message content. This data feeds into the respective AI Agent n8n node (Gmail or Telegram).

AI Agent Execution: Both Agent nodes are configured with:
An OpenAI Chat Model for processing and generation.
A Memory Buffer Window n8n node to maintain session context (essential for multi-turn conversations).
Google Sheets Tool n8n nodes for RAG (Retrieval-Augmented Generation) access to the knowledge base and for logging escalation requirements.

The AI generates a contextual and helpful response. If the inquiry is complex, the agent logs the data to the escalation sheet.

Final Actions:
Gmail: The workflow uses the Gmail n8n node to send the AI-generated reply, marks the original email as read, and logs the interaction details (message, response, timestamp) to a dedicated log sheet in Google Sheets.

• Telegram: The workflow uses the Telegram n8n node to send the reply directly to the customer and logs the interaction to the same Google Sheets log.

To deploy this comprehensive n8n workflow template, follow these setup steps:

1. Import: Copy the provided JSON data and paste it into your n8n instance to create a new workflow.


2. OpenAI Credentials: Locate the 🧠 OpenAI Chat Model n8n nodes and set up your OpenAI API key credentials.


3. Gmail Setup: Configure the 📧 Gmail Customer Inquiry Trigger and 📧 Send Gmail Reply n8n nodes with your OAuth2 Gmail account credentials. Ensure the account has permissions to read and send messages.


4. Telegram Setup: Configure the 💬 Telegram Customer Message Trigger and 💬 Send Telegram Reply n8n nodes using your Telegram bot API key credentials.


5. Google Sheets Setup: Update all Google Sheets n8n nodes (knowledge base, Escalation Logger, Log Interaction) with your Google Sheets OAuth2 credentials. You must replace the placeholder Document ID and Sheet Names with those corresponding to your knowledge base and logging spreadsheets.


6. Activation: Once all credentials and configurations are updated, test the workflow manually and then activate the n8n workflow to run automatically.

This n8n workflow leverages powerful Langchain integrations alongside core n8n nodes for complex routing and data handling:

Gmail Trigger: The primary n8n trigger node for email, set to poll for new unread messages categorized as personal.
Telegram Trigger: The dedicated n8n trigger node for messaging, set to trigger on incoming message updates.
@n8n/n8n-nodes-langchain.agent (AI Agents): These crucial n8n nodes define the core logic and personality of the support system. They are instructed via detailed system messages to use the available tools (Google Sheets knowledge base) to generate professional, context-aware responses.
@n8n/n8n-nodes-langchain.memoryBufferWindow (Memory Nodes): These n8n nodes are vital for maintaining conversation state, keying the session memory using the customer's email or Telegram ID.
Google Sheets Tool (knowledge base): Configured as a tool within the AI Agents, allowing the GPT model to query external, current information (your FAQ/Knowledge Base sheet) before formulating a reply.
Google Sheets (Log Interaction): Standard n8n node used to append a record of the entire interaction (customer query, AI response, channel, time) for audit purposes.

• Error Trigger: This advanced n8n trigger captures any execution failures and routes the error data to the 📢 Notify Admin of Error node for immediate alerting.