AI Lead Qualification Chatbot
A webhook-driven AI agent that qualifies inbound leads over SMS or web chat, remembers each conversation separately, and hands off to a human at the right moment.
The problem
Inbound leads ask the same five questions, and answering them consumes a salesperson's entire morning. Generic chatbots make it worse by inventing pricing and availability, which you then have to walk back.
What this workflow does
An AI agent works through need, timeline and decision authority one question at a time, with a per-contact memory buffer so each lead keeps their own thread. The system prompt explicitly forbids inventing pricing, guarantees or hours, and anything it does not know is escalated. When the lead qualifies or asks for a human, sales gets a Slack notification with the conversation context.
How it works
- Your SMS or web chat provider posts inbound messages to the webhook.
- The agent loads the last 12 turns of that specific contact's conversation from memory.
- It asks one qualifying question at a time and returns structured JSON alongside the reply.
- A Code node parses that JSON defensively and falls back to raw text if the model drifts.
- The reply is sent through your CRM, and qualified or escalating conversations ping sales in Slack.
What it connects
- n8n
- OpenAI
- GoHighLevel
- Slack