LeadGen.AI

Published: May 01, 2025

Overview

LeadGen.AI is an AI-powered lead generation and automated outreach platform designed for B2B sales teams. The system intelligently identifies qualified prospects by analyzing industry events, company profiles, and stakeholder data, then generates personalized outreach messages at scale.

Originally built for DuPont Tedlar’s Graphics & Signage team, the platform is architected with scalability and generalizability in mind - ready to be adapted across different product teams and industries.

LeadGen.AI Platform Interface

Problem Statement

Traditional B2B lead generation is time-consuming and inefficient:

Sales teams manually research industry events and trade shows
Identifying decision-makers at target companies requires extensive networking
Creating personalized outreach messages for each prospect is labor-intensive
Qualifying leads against ideal customer profiles happens inconsistently

LeadGen.AI automates this entire workflow while maintaining high personalization quality.

Demo Video

Key Features

🎯 Ideal Customer Profile (ICP) Management

Define and save multiple ICPs with industry, revenue, size, and geographic criteria
Reuse profiles across different campaigns
Track performance metrics for each ICP

🔍 Event-Based Lead Discovery

Live web scraping of industry events (FESPA, trade shows, conferences)
Source citation and verification for all event data
Exhibitor identification and company extraction
Custom CSV upload for proprietary company lists

📋 Stakeholder Persona Targeting

Define target roles: Decision Makers, Influencers, End Users
Job title and department filtering
LinkedIn Sales Navigator integration (in development)
Multi-persona targeting within organizations

🤖 AI-Powered Lead Qualification

Strategic relevance scoring based on custom criteria
Company size, revenue, and growth stage filtering
Technology stack and funding status filters
Match percentage calculation with detailed reasoning

💼 Intelligent Data Enrichment

Company profile enhancement (technologies, funding, news)
Stakeholder contact information discovery
Competitor analysis
Industry engagement metrics

✉️ Personalized Outreach Generation

Context-aware message templates
ICP and persona-specific customization
Multiple channel support (email, LinkedIn)
A/B testing capabilities

📊 Lead Management & Export

Sortable, filterable lead tables
Status tracking (New, Contacted, Qualified, Disqualified)
CSV and JSON export formats
CRM integration ready (in development)

Technical Architecture

Tech Stack

Frontend:

React with TypeScript
Vite for build tooling
Tailwind CSS for styling
shadcn/ui component library

Backend:

Node.js/Express server
PostgreSQL database with Drizzle ORM
OpenAI GPT-4 for AI generation
Web scraping with Cheerio/Puppeteer

Data Flow

Input Layer: User defines ICP and selects target events
Discovery Layer: System scrapes event exhibitors and company data
Processing Layer: AI analyzes companies against ICP criteria
Qualification Layer: Scoring, filtering, and ranking of leads
Enrichment Layer: Additional data gathering for qualified leads
Output Layer: Personalized messages and exportable lead lists

User Workflow

Step 1: ICP Selection/Creation

Browse existing ICPs or create new profiles
Define industry, size, revenue, geography, and custom criteria
Save for future campaigns

Step 2: Target Event Selection

Browse live-scraped industry events with exhibitor counts
View source citations for transparency
Sync latest events with one click
Alternative: Upload custom company lists via CSV

Step 3: Configure Filters

Stakeholder Personas: Define target roles and titles
Strategic Qualification: Set company size, revenue thresholds
Relevance Criteria: Industry engagement, market activity, innovation focus
Additional Filters: Technology stack, funding status, keywords
Generation Options: Lead count, enrichment toggle, message generation

Step 4: Identify Decision Makers

LinkedIn Sales Navigator search (in development)
Job title and company filtering
Relevance scoring with rationale
Contact information discovery

Step 5: Generate & Review Leads

AI analyzes event data against ICP
Each lead includes:
- Company profile and stakeholder details
- Match percentage with detailed reasons
- Enrichment data (technologies, funding, news)
- Personalized outreach message
Filter, sort, and update lead status

Step 6: Export & Integrate

Export to CSV or JSON
Select fields and filtering options
Import into CRM systems
Begin outreach campaigns

Implementation Highlights

AI-Powered Lead Scoring

The system uses GPT-4 to analyze companies against multiple criteria:

const qualificationPrompt = `
Analyze this company against the following ICP criteria:
- Industry: ${icp.industry}
- Revenue Range: ${icp.minRevenue} - ${icp.maxRevenue}
- Employee Count: ${icp.minEmployees} - ${icp.maxEmployees}
- Strategic Criteria: ${strategicFilters}

Company Data:
${companyProfile}

Provide:
1. Match score (0-100)
2. Specific reasons for the score
3. Key alignment factors
4. Potential concerns
`;

Web Scraping Pipeline

Live event data extraction with citation tracking:

async function scrapeEvents() {
  const events = await scraper.fetch('https://fespa.com/events');
  
  return events.map(event => ({
    name: event.title,
    date: event.date,
    location: event.location,
    exhibitorCount: event.exhibitors.length,
    sourceUrl: event.url,
    lastUpdated: new Date()
  }));
}

Personalized Outreach Generation

Context-aware message creation:

const messagePrompt = `
Create a personalized outreach message for:

Recipient: ${stakeholder.name}, ${stakeholder.title}
Company: ${company.name}
Context: Met at ${event.name}

ICP Focus: ${icp.industry}
Key Value Props: ${icp.valuePropositions}

Match Reasons: ${lead.matchReasons}

Generate a professional, concise message that:
1. References the event context
2. Highlights relevant value propositions
3. Includes a clear call-to-action
`;

Results & Impact

Time Savings: Reduces lead research time from hours to minutes
Personalization at Scale: Generates hundreds of customized messages automatically
Quality Improvement: AI scoring ensures consistent lead qualification
Data Transparency: Source citations build trust in lead quality
Workflow Integration: Export capabilities enable seamless CRM adoption

Future Enhancements

Near-Term Improvements

Multiple Event Selection: Select and analyze multiple events simultaneously
Progress Indicators: Real-time generation progress bars and animations
CRM Integration: Direct export to Salesforce, HubSpot, and other platforms
Match Detail Transparency: Detailed criteria matching with source attribution

Long-Term Vision

LinkedIn API Integration: Full Sales Navigator integration for richer stakeholder data
Feedback Loop: User interaction tracking to improve AI recommendations
Multi-Channel Outreach: Email, LinkedIn, and call script generation
Analytics Dashboard: Campaign performance tracking and ROI metrics
Team Collaboration: Shared ICPs, lead assignments, and activity tracking
Industry Expansion: Pre-built ICP templates for various verticals

Technical Challenges Solved

Challenge 1: Dynamic Web Scraping

Events websites frequently change structure. Solution: Flexible scraping patterns with fallback strategies.

Challenge 2: Data Quality

Inconsistent company information across sources. Solution: Multi-source validation and confidence scoring.

Challenge 3: AI Consistency

GPT outputs can vary. Solution: Structured prompts with JSON schema enforcement and temperature tuning.

Challenge 4: Scale & Performance

Processing hundreds of companies efficiently. Solution: Batch processing, caching, and async operations.

Installation & Setup

# Clone repository
git clone https://github.com/nicolehao34/LeadGen.AI.git
cd LeadGen.AI

# Install dependencies
npm install

# Set environment variables
export OPENAI_API_KEY=your_key_here

# Start development server
npm run dev

# Access at http://localhost:3000

Nicole Hao