retention-analyzer
Predict churn, identify at-risk customers, and automate retention campaigns
SKILL.mdskills/retention-analyzer/SKILL.md
GitHub🔄 Retention Analyzer Agent
Purpose: Stop customers from leaving before they do. Predict churn, trigger retention campaigns, and maximize lifetime value.
When to Use
# Analyze churn risk
@retention-analyzer risk --segment all
# Identify at-risk customers
@retention-analyzer identify --threshold high
# Trigger retention campaign
@retention-analyzer campaign --risk-level medium
# View retention metrics
@retention-analyzer metrics --period 30d
# Predict future churn
@retention-analyzer predict --months 3
Capabilities
1. Churn Prediction
- Behavioral scoring (login frequency, usage patterns)
- Engagement decline detection
- Support ticket sentiment analysis
- Payment failure tracking
- ML-powered risk scoring
2. At-Risk Identification
- Segment customers by churn probability
- Priority ranking (who to save first)
- Revenue impact calculation
- Root cause analysis
- Early warning alerts
3. Automated Retention
- Triggered email sequences
- Personalized win-back offers
- Proactive support outreach
- Loyalty reward automation
- Exit interview requests
4. Retention Analytics
- Cohort retention curves
- LTV predictions
- Churn rate by segment
- Campaign effectiveness
- ROI of retention efforts
Usage Examples
Churn Risk Analysis
User: @retention-analyzer risk
Agent: ⚠️ CHURN RISK ANALYSIS
Total Customers: 12,450
RISK BREAKDOWN:
🔴 High Risk (Churn >70%): 487 customers (3.9%)
🟡 Medium Risk (Churn 30-70%): 1,243 customers (10.0%)
🟢 Low Risk (Churn <30%): 10,720 customers (86.1%)
HIGH RISK INDICATORS:
• No login in 30+ days: 312 customers
• Failed payment 2+ times: 89 customers
• Support ticket unresolved: 47 customers
• Usage dropped 80%+: 39 customers
REVENUE AT RISK: $48,700/month
TOP 10 CUSTOMERS TO SAVE:
1. Acme Corp ($2,400/mo) - No login 45 days
2. TechStart Inc ($1,800/mo) - Payment failed 3x
3. GrowthLabs ($1,500/mo) - Support escalation
[7 more...]
IMMEDIATE ACTION REQUIRED ⚡
Retention Campaign
User: @retention-analyzer campaign --risk-level high
Agent: 🎯 RETENTION CAMPAIGN ACTIVATED
Target: 487 high-risk customers
Strategy: Multi-channel intervention
SEQUENCE 1: Email (Day 0)
Subject: "We miss you - here's 20% off to come back"
Hi [Name],
We noticed you haven't logged in recently. Everything okay?
We've been busy improving Brand Marketing Team:
✨ New feature: AI video scripts
✨ Faster: 2x speed improvements
✨ Better: 15 new marketing templates
Come back and get 20% off your next 3 months.
[Claim Discount →]
We're here if you need help.
Best,
The Team
SEQUENCE 2: Personal Outreach (Day 3, if no response)
• Account manager phone call
• Calendly link for demo
• "What would bring you back?" survey
SEQUENCE 3: Final Offer (Day 7)
Subject: "Last chance: 30% off + free onboarding"
[More aggressive offer]
Campaign launched ✓
Tracking responses... ✓
Expected save rate: 15-25%
Configuration
retention-analyzer:
scoring:
login-weight: 30
usage-weight: 25
support-weight: 20
payment-weight: 25
thresholds:
high: 70
medium: 30
campaigns:
high-risk: "aggressive-winback"
medium-risk: "engagement-boost"
low-risk: "loyalty-rewards"
Part of the Brand Marketing Team framework. Author: @namanwtf | Version 3.0.0 | MIT License