Skill: Connection Strategy Manager

Purpose

Manage LinkedIn connection request targeting and execution per tier with intelligent note selection, daily limit enforcement, and human-like distribution patterns.

What it does

• •Gets connection targets for a profile filtered by tier and daily limit
• •Executes connection requests with optional notes
• •Applies tier-specific note strategies (70% no note, 20% generic, 10% personalized)
• •Tracks note distribution and acceptance metrics
• •Logs all requests for audit trail

Why it matters

Connection requests build the foundation for future opportunities. But LinkedIn throttles connections if sent too fast or mechanically. By applying tier-specific targeting and note strategies, we build networks that look natural while respecting daily limits.

How to use it

Get connection targets for today

from connection_manager.connection_strategy import get_connection_targets

targets = get_connection_targets({
    "profile_id": "P-001",
    "tier": "A",
    "date": "2026-02-15"
})

print(f"Available targets: {targets['targets_available']}")
print(f"Daily limit: {targets['daily_limit']}")
for target in targets['targets']:
    print(f"  - {target['name']} ({target['title']}) at {target['company']}")

Execute single connection request

from connection_manager.connection_strategy import execute_connection_request

result = execute_connection_request({
    "profile_id": "P-001",
    "target_id": "T-5678",
    "target_name": "Jane Recruiter",
    "target_title": "Senior Technical Recruiter",
    "target_company": "TechStaff Inc",
    "note": "Hi! I'd like to connect.",
    "proxy_ip": "192.0.2.15"
})

print(f"Connection request {result['request_id']} sent")

Execute batch connections

from connection_manager.connection_strategy import execute_batch_connections

batch = execute_batch_connections({
    "profile_id": "P-001",
    "tier": "A",
    "date": "2026-02-15",
    "targets": targets_list,
    "proxy_ip": "192.0.2.15"
})

print(f"Sent {batch['total_requests']} connection requests")
print(f"Notes: {batch['statistics']['note_distribution']}")

Daily connection limits

Target distribution strategies

Tier A (Thoroughbred)

Focus: High-quality recruiter and hiring manager targeting

Note Strategy: 70% no note, 20% generic, 10% personalized

Tier B (Growth)

Focus: Mixed recruiter and HR professional targeting

Note Strategy: 70% no note, 25% generic, 5% personalized

Tier C (Warming)

Focus: Network building with low-pressure approach

Note Strategy: 80% no note, 15% generic, 5% personalized

Tier D (Dormant)

Focus: No activity

Note strategies

No Note (70% for Tier A, 80% for Tier C)

• •Just the connection request without message
• •Fastest to send
• •Good for open networkers and established networkers
• •Most natural for large-scale connection building

Generic Note (20% for Tier A, 15% for Tier C)

• •Short, professional message
• •Examples:
  • •"Hi! I'd like to connect and expand my professional network."
  • •"Hi! Let's connect and share industry insights."
  • •"Would love to grow my network. Let's connect!"
• •Shows intent without being too specific
• •Works for recruiters and HR

Personalized Note (10% for Tier A, 5% for Tier C)

• •References target's company or background
• •Examples:
  • •"Hi! I noticed your work at TechStaff and would love to connect."
  • •"Hi! As a Senior Recruiter, I think we'd have great insights to share."
  • •"Hi! I'm impressed with your background. Let's connect!"
• •Higher acceptance rate but slower to execute
• •Reserved for high-value targets (Tier 1 recruiters, hiring managers)

Input/Output schemas

Get connection targets

Input:

{
  "profile_id": "P-001",
  "tier": "A",
  "date": "2026-02-15",
  "daily_limit": 20
}

Output:

{
  "profile_id": "P-001",
  "date": "2026-02-15",
  "tier": "A",
  "tier_name": "Thoroughbred",
  "daily_limit": 20,
  "targets_available": 20,
  "targets": [
    {
      "target_id": "T-512847",
      "category": "tier1_recruiters",
      "name": "Sarah Johnson",
      "title": "Senior Technical Recruiter",
      "company": "Cognizant",
      "connection_count": 1250
    }
  ],
  "target_distribution": {
    "tier1_recruiters": 0.40,
    "tier2_recruiters": 0.30,
    "hiring_managers": 0.20,
    "industry_peers": 0.10
  },
  "note_strategy": {
    "no_note": 0.70,
    "generic_note": 0.20,
    "personalized_note": 0.10
  }
}

Execute connection request

Input:

{
  "profile_id": "P-001",
  "target_id": "T-5678",
  "target_name": "Jane Recruiter",
  "target_title": "Senior Technical Recruiter",
  "target_company": "TechStaff Inc",
  "note": "Hi! I noticed your work at TechStaff and would love to connect.",
  "proxy_ip": "192.0.2.15"
}

Output:

{
  "request_id": "CONN-REQ-20260215103245-45678",
  "profile_id": "P-001",
  "target_id": "T-5678",
  "target_name": "Jane Recruiter",
  "target_title": "Senior Technical Recruiter",
  "status": "SENT",
  "note_sent": true,
  "note_type": "personalized",
  "proxy_ip": "192.0.2.15",
  "timestamp": "2026-02-15T10:32:45Z"
}

Execute batch connections

Input:

{
  "profile_id": "P-001",
  "tier": "A",
  "date": "2026-02-15",
  "targets": [...],
  "proxy_ip": "192.0.2.15"
}

Output:

{
  "profile_id": "P-001",
  "date": "2026-02-15",
  "tier": "A",
  "total_requests": 20,
  "requests": [...],
  "statistics": {
    "note_distribution": {
      "none": 14,
      "generic": 4,
      "personalized": 2
    },
    "no_note_percentage": 70.0,
    "with_note_percentage": 30.0
  },
  "timestamp": "2026-02-15T10:45:12Z"
}

Performance

• •Get targets: <50ms
• •Single connection request: <5ms
• •Batch of 20: <100ms

Integration points

• •Schedule Generator: Tells when to send connections
• •Activity Simulator: Connection requests are scheduled as part of daily activity
• •Portfolio Manager: Tracks total connections sent per day
• •Learning System: Connection acceptance rates feed into profile health

Metrics tracked

• •Connections sent per profile per day
• •Connection acceptance rate (leading indicator of account health)
• •Note effectiveness (acceptance rate by note type)
• •Target category performance (which categories have best acceptance)
• •Regional patterns (if tracking by timezone)

Safety constraints

• •Daily limit enforced per tier
• •No more than 20 per day per profile (LinkedIn hard limit)
• •Time gaps between requests (randomized 3-15 seconds)
• •IP not reused across profiles same day
• •Human-like note distribution (not all generic, not all personalized)

Future enhancements

• •Machine learning to predict acceptance rates per target
• •Budget-aware targeting (focus on high-value targets when nearing limit)
• •Relationship history tracking (don't re-request if already connected)
• •A/B testing of notes (track which performs better)
• •Temporal patterns (when do acceptances happen)