service-business-prospecting（服务业勘探）

Service Business Prospecting

Overview

This skill runs a daily 6am lead-generation job for Ventured Brands, who sells AI automation (missed-call text-back, AI receptionist, review automation, full-stack bundles) to local US service businesses.

Objective

Produce a fresh batch of ~25 qualified, contactable leads in one US metro for one of the four target industries, saved as a daily-dated CSV that drops cleanly into the AI_Automation_Lead_Tracker.xlsx.

Target Industries (4)

Pitch Angles (4)

ICP (Ideal Customer Profile)

Target:

• Local US service businesses (not chains/franchises)
• 2-10 employees
• Established (2+ years in business)
• Has Google/Yelp presence
• Not already using AI competitors

NOT Target:

• Chains (Servicemaster, etc.)
• Solo operators
• New businesses (<1 year)
• Enterprise/commercial only

Day Rotation

Scoring Rubric

Workflow (9 Steps)

Step 1: Check Rotation

• Read today's day-of-week
• Get industry-city combo from rotation table
• Check last 7 days of output to avoid repeats

Step 2: WebSearch - Business Discovery

Search queries per industry:

• "[Industry type] [City] AZ phone number address"
• "[Industry type] [City] AZ owner name email"
• "[Industry type] [City] AZ yelp reviews"

Get minimum 10 real businesses. NEVER fabricate.

Step 3: ICP Filtering

Remove:

• Chains/franchises (name includes "Mister", "Merry", "ServiceMaster", "College")
• New businesses (<2 years)
• Commercial-only

Step 4: Contact Enrichment

For each lead:

• Extract phone number (primary)
• Find email from website/Facebook
• Get owner/contact name
• Verify address

Step 5: Scoring

Assign Tier A/B/C/D based on contact completeness.

Step 6: Pitch Assignment

Match to best pitch angle:

• Has receptionist/front desk → AI Receptionist
• Has reviews → Review Automation
• Small biz → Missed-Call Text-Back
• Enterprise-ready → Full-Stack Bundle

Step 7: Build Output

Create CSV with columns:

Company Name, Owner/Contact, Phone, Email, Address, City, State, Zip, Industry, Pitch Angle, Tier, Status, Last Contacted, Notes, Source

Step 8: Quality Check

• Verify 10+ leads have phones
• Report email yield
• Flag any data gaps

Step 9: Save Output

Save to:

<outputs-folder>/daily-leads/<YYYY-MM-DD>-<industry-slug>-<city-slug>.csv

Also save notes.md with summary.

Data Quality Rules

• NEVER fabricate businesses, owner names, phone numbers, or emails
• If WebSearch returns <10 real leads, report the shortfall
• All leads must be real, verifiable businesses
• Email must be from website/Facebook (not guessed)

Tracker Vocabulary (Reference)

Industry values:

• "HVAC"
• "Plumbing"
• "Electrical"
• "Med Spa"
• "Dental"
• "Chiro"
• "Auto Detail"
• "Auto Repair"
• "Cleaning"
• "Landscaping"
• "Pest Control"

Pitch Angle values:

• "Missed-Call Text-Back"
• "AI Receptionist"
• "Review Automation"
• "Full-Stack Bundle"

Status values:

• "Not Contacted"
• "Attempted"
• "Connected"
• "Qualified"
• "Proposal Sent"
• "Closed Won"
• "Closed Lost"

Output Files

1. CSV - 2026-04-28-med-spa-dental-chiro-scottsdale.csv
2. Notes - notes.md with:

- Industry+City used - Sources checked - Count by tier - Email yield - Data quality caveats