Auto-respond and classify WhatsApp leads with Ollama AI and log to Google Sheets

Who is this for

This workflow is built for sales teams, agencies, and small businesses that receive inbound leads via WhatsApp and want to automate their first response, lead qualification, and CRM logging — without missing a single message.

What this workflow does

It listens for incoming WhatsApp messages, uses an AI agent to classify each message by intent (hot lead, warm lead, support, or needs qualification), sends a tailored auto-reply, logs every interaction to Google Sheets, and automatically books Google Calendar meetings with Meet links for qualified leads.

How it works

• WhatsApp Trigger receives incoming messages and filters out bot/status messages to prevent loops.
• AI Agent (powered by Ollama or any connected LLM) classifies the message into one of four intent categories with confidence scoring.
• Smart Router directs each intent down a dedicated path.
• Hot & Warm Leads receive an instant reply, get logged to Google Sheets, have a Google Calendar meeting auto-booked, and receive the Meet link via WhatsApp.
• Support requests are logged and receive a ticket confirmation.
• Vague or incomplete messages trigger a smart follow-up question. Conversation memory ensures the AI re-classifies correctly when the user replies with more context.

Setup steps

• Connect your WhatsApp Business API credentials (Meta Cloud API).
• Connect Google Sheets OAuth and set your spreadsheet ID in all three logging nodes.
• Connect Google Calendar OAuth and select your calendar in both booking nodes.
• Configure your LLM (Ollama endpoint, OpenAI, or any supported model).
• Update the BOT_NUMBERS array in the "Parse WhatsApp Message" node to match your WhatsApp Business phone number ID.
• Update the phoneNumberId in all WhatsApp Send nodes to your number.
• Send a test message and verify the full flow.

Requirements

• WhatsApp Business API (Meta Cloud API) access
• Google Sheets and Google Calendar accounts with OAuth credentials
• An LLM endpoint (Ollama, OpenAI, or any n8n-supported model)
• n8n instance (cloud or self-hosted)

How to customize

• Swap the AI model in the Ollama Chat Model node for OpenAI, Anthropic, or any supported LLM.
• Edit the auto-reply templates in each Reply code node to match your brand voice.
• Adjust meeting booking times (default: Hot = 2 hours out, Warm = 4 hours out).
• Add Slack or email notifications by branching from the Google Sheets logging nodes.
• Modify the AI classification prompt to add custom intent categories for your business.