Project Overview

AI Sales Ecosystem is an end-to-end AI sales system for the B2B company GrowthLabs: from the first contact with a lead to the weekly report for management. The system qualifies incoming leads using the BANT model (through a website chatbot or a voice AI agent), creates a CRM record with a score, analyses dialogue quality daily and every Friday sends the CEO a full funnel report in Telegram.

Three workflows operate as one closed loop: qualification, dialogue analysis and management analytics — with no manual work from the sales team.

Context

GrowthLabs is a B2B company that receives inbound leads from its website and handles them with a small sales team. Every lead is a potential deal, but they are far from equally ready to buy: one wants a demo right now, another is just checking prices, a third doesn't fit the budget.

Management wanted three things: incoming leads qualified automatically, dialogue quality reviewed regularly, and a weekly funnel picture for the CEO without anyone assembling numbers by hand.

Problem

The main problems before automation:

Solution

A closed AI ecosystem built from three workflows in n8n.

The first workflow qualifies a lead using BANT — and works in two formats: an AI chatbot on the website, or a voice AI agent that calls the lead directly. The outcome is the same: the lead gets a score and lands in the CRM as a ready-to-work record.

The second workflow analyses dialogues daily — both chat and voice — identifying agent mistakes, the customer's real problem and successful patterns.

The third workflow runs every Friday: it takes funnel metrics from the CRM plus the dialogue analysis and delivers a structured report to the CEO in Telegram.

Process

How the system is built — step by step: