Building a sales call list by hand is one of those soul-crushing tasks that every B2B rep knows too well. Open Google Maps. Search "roofing companies near me." Click into each listing. Open the website in a new tab. Hunt for a phone number buried somewhere on the page—assuming it's not hidden behind a "Contact Us" form that requires an email to unlock. Copy it into a spreadsheet. Repeat two hundred times. By hour six, you're questioning your career choices and your will to live.

The Stack That Changes Everything

A developer going by girma35 on DEV.to documented their solution: combining Claude AI with Apify's web scraping infrastructure and the Model Context Protocol (MCP) to fully automate lead list generation. The architecture uses Apify actors—pre-built web crawlers optimized for sites like Google Maps—to extract business listings at scale. Instead of manually clicking through hundreds of pages, a single API call kicks off an automated crawl that gathers business names, addresses, and any publicly visible contact data.

Where Claude Fits In

The real magic happens when Claude AI processes the raw scraped data. The model handles natural language extraction tasks that would require fragile regex patterns or endless edge-case handling—like identifying phone numbers buried in JavaScript-rendered pages, parsing semi-structured business listings, and flagging entries with missing information for follow-up scraping passes. MCP serves as the connective tissue, allowing Claude to communicate directly with Apify's API endpoints without custom integration code.

Real Results for Sales Teams

According to the author's estimates, this pipeline eliminated roughly 40 hours of manual prospecting work per month for their team. The setup handles geographic targeting through Google Maps parameters, automatically enriches records with website data when available, and outputs clean CSV files ready for CRM import. For sales ops teams drowning in list-building tasks, this represents a genuine productivity unlock—no more copy-pasting phone numbers until your wrists ache.

Key Takeaways

  • Apify actors provide battle-tested scrapers that handle Google Maps' anti-bot measures out of the box
  • Claude's instruction-following capabilities eliminate brittle parsing logic for messy real-world data
  • MCP enables direct agent-to-API communication without custom webhook infrastructure
  • The full pipeline runs on scheduled triggers, delivering fresh lead lists automatically each week

The Bottom Line

This isn't theoretical AI hype—it's a practical automation stack that any developer could build in an afternoon. If your sales team is still building prospect lists manually, you're burning money. Full stop.