Lead Pipeline — Automated B2B Client Acquisition

Automates B2B lead generation and outreach: scrapes leads, enriches them with company data, scores by fit, generates personalized cold emails using AI, and sends them via SMTP.

Features

• Multi-source lead collection — web scraping + LinkedIn (via Proxycurl / Apollo / mock)
• Lead enrichment — company size, tech stack, revenue range (Clearbit optional)
• Scoring engine — 0–100 score based on title seniority, email quality, company fit
• AI-generated messages — personalized cold emails via Claude API
• SMTP sending — with open tracking pixel + unsubscribe footer
• SQLite storage — deduplication, status tracking, export to CSV
• Scheduler — runs automatically on a configurable interval
• CLI — full control from the terminal

Quick Start

# 1. Install dependencies
pip install -r requirements.txt

# 2. Configure
cp .env.example .env
# Fill in ANTHROPIC_API_KEY, SMTP_USER, SMTP_PASS, SENDER_NAME, VALUE_PROPOSITION

# 3. Dry run (collect leads, no emails sent)
python main.py run --niche "ecommerce Poland" --dry-run

# 4. Full run
python main.py run --niche "SaaS" --limit 30

# 5. Schedule (every 24h)
python main.py schedule --interval 24 --niche "SaaS"

# 6. Report
python main.py report --last 7

Architecture

main.py                      # CLI entry point
src/
  pipeline.py                # Orchestrates the full workflow
  scrapers/
    web_scraper.py           # Extracts emails from websites
    linkedin_scraper.py      # LinkedIn via Proxycurl / Apollo / mock
  outreach/
    message_generator.py     # Personalized cold emails via Claude API
    email_sender.py          # SMTP sender with tracking pixel
  storage/
    models.py                # Lead dataclass + LeadStatus enum
    lead_store.py            # SQLite persistence + deduplication
  utils/
    enricher.py              # Lead enrichment + scoring (0-100)
    config.py                # YAML + .env config loader
    logger.py                # File + console logging
    reporter.py              # Funnel analytics
config/
  config.yaml                # Non-sensitive configuration
tests/
  test_pipeline.py           # pytest test suite
.github/
  workflows/ci.yml           # GitHub Actions CI (Python 3.11-3.13)

LinkedIn Providers

Provider	Setup
mock	No setup needed — synthetic leads for dev/testing
apollo	Set APOLLO_API_KEY
proxycurl	Set PROXYCURL_API_KEY

Lead Scoring

Signal	Points
Business email	+15
Name present	+10
Company present	+10
LinkedIn URL	+10
CEO/Founder title	+20
C-suite / Director	+15
Company size 10–200	+15
Valuable tech stack	+5 per match

Tests

pytest tests/ -v

License

MIT