🔧 AI-Powered Cold Call Machine

🎯 Purpose

The AI-Powered Cold Call Machine is a fully automated workflow designed to generate qualified leads from LinkedIn, evaluate them using AI-based scoring, identify key decision-makers, and generate personalized cold call scripts. All results are saved to a Google Sheet-based CRM.

⚙️ How It Works

1. Initialization

• Triggered either manually or via schedule.
• Pulls configuration from a Google Sheet’s Settings tab (e.g., target product, keywords, company size, API key).

2. Company Search on LinkedIn

• Uses the Ghost Genius API to search for companies based on cleaned, relevant keywords extracted by OpenAI.
• Handles pagination, up to 1000 companies per batch.

3. Company Filtering

Each company goes through:

• Data enrichment via Ghost Genius (website, size, followers, etc.).
• Filtering:
  • Must have a LinkedIn page with a website.
  • Must have 200+ followers.
• Deduplication: checks if the company already exists in the CRM.

4. AI-Based Company Scoring

• A specialized AI model scores each company from 0 to 10 based on:
  • Industry fit.
  • Size/location alignment.
  • Potential pain points that match your offering.
• If the company is new and relevant (score ≥ 7), it is saved in the Companies sheet.

5. Decision Maker Identification

• Uses Sales Navigator API (via Ghost Genius) to find employees with targeted job titles.

• For each matching profile:

  • Enriches contact data (title, bio, etc.).
  • Retrieves phone number (if available).
  • Generates a 20-second personalized cold call script using OpenAI, based on company and profile data.
  • Saves all information in the Leads tab of the CRM.

• If no decision maker is found, the company status is marked accordingly.

📈 Outcome

• A fully enriched, qualified lead database.
• Custom cold call scripts ready to be used by SDRs or founders.
• Zero manual work – from search to lead generation, everything is automated.

💡 Use Case

Perfect for SDRs, founders, or growth marketers looking to scale cold outreach without sacrificing personalization or running into LinkedIn scraping limits.