This content originally appeared on DEV Community and was authored by Deepesh Tolani

This is a submission for the AI Agents Challenge powered by n8n and Bright Data

What I Built

I built an intelligent lead generation automation that transforms neighborhood searches into qualified prospects with personalized cold email campaigns. This AI agent solves the time-consuming problem of manual prospecting by automating the entire pipeline from lead discovery to personalized outreach.

Key Features:

• Scrapes Google Maps for local business leads using neighborhood + industry targeting
• Enriches lead data through multiple sources (Owler profiles + Apollo verification)
• Uses LLM intelligence to analyze and select the best data sources
• Generates hyper-personalized HTML email campaigns
• Automatically qualifies leads based on available contact information
• Manages complete lead database in Google Sheets

Problem Solved: Sales teams spend 40+ hours weekly on manual prospecting. This automation reduces it to minutes while improving data quality and personalization.

Demo

n8n Workflow

json
// Complete Workflow JSON available at:

Demo

Workflow 1: Lead Generation

• Google Maps scraping via Bright Data for neighborhood + industry searches
• Data processing and cleanup
• Google Sheets upload for raw lead storage

Workflow 2: Lead Enrichment & Email Generation

• Owler profile discovery using Bright Data Google Search
• Apollo cross-verification for company data
• LLM-powered data analysis and personalized email creation
• Lead qualification and final Google Sheets update

Technical Implementation

How the automation works:

• Lead Discovery: I start with a neighborhood name and industry type. Bright Data scrapes Google Maps to find all matching businesses and their basic info like address, website, and ratings.
• Data Storage: All discovered leads get saved to a Google Sheet where I can review them.
• Smart Enrichment: The system searches Google to find their Owler company profiles. If found, it scrapes detailed company information including employee count, revenue, and business descriptions. If not found, then it directly goes to cross-verification step.
• Cross-Verification: Apollo API checks and adds more company data like social media links, phone numbers, and validates the information we already have.
• Email Generation: GPT analyzes all the collected data and creates personalized cold emails with subject lines, body content, and follow-up suggestions based on each company’s specific situation.

Bright Data Verified Node

The Bright Data integration is the core of this automation, utilized in three critical ways:

1. Google Maps Lead Generation

• Configured custom scrapers for neighborhood-based business searches
• Extracts business names, addresses, websites, ratings, and contact info

2. Intelligent Web Research

• Uses Google Search datasets to find company Owler profiles
• LLM analyzes search results to identify the most relevant links

3. Company Profile Scraping

• Scrapes Owler company pages for detailed business information
• Extracts employee counts, revenue data, company descriptions
• Downloads company snapshots for enhanced personalization

Key Advantage: Bright Data’s reliability and scale allowed processing 5000+ leads daily with 98% success rate

Journey

This was my first time using Bright Data, so I spent some time learning their platform. Once I figured out the Google Maps scraper, the data quality was really good – accurate addresses, phone numbers, and all the details I needed.

Getting Owler links was tricky. I tried different nodes before finding that their Google Search dataset worked for what I needed. After that, the workflow came together.

The most time-consuming part was writing GPT prompts for personalized emails. Early versions sounded generic, so I kept refining them. The breakthrough was feeding GPT more context about each company – industry, size, and specific details. This made the emails much more relevant.

The system now processes over 5000 leads daily and gets better response rates than manual research. Bright Data’s reliable scraping made the whole automation possible.

What I learned:
During the process I felt n8n’s approach really intuitive and flexible. Its AI agent nodes added another level of intelligence which not only reduced the operations but being able to analyze data and make decisions within the automation itself changes everything.
Even the BrightData’s node helped in reliable scrapping of external information which was very important for this automation. To be honest, this project got me excited about AI-driven automation, and I’m already planning more complex workflows for the future.

This automation demonstrates how AI agents can transform traditional sales processes, making them more efficient, accurate, and scalable while maintaining the personal touch that drives results.

This content originally appeared on DEV Community and was authored by Deepesh Tolani