| name | churn-prevention-playbook |
| description | Creates churn prevention playbooks with early warning indicators, intervention sequences, and save offer strategies. Use when reducing customer churn in subscription businesses. |
| allowed-tools | Read Write Glob |
| metadata | {"author":"matthewhitcham","version":"1.0"} |
Churn Prevention Playbook
When to Use This Skill
Use this skill when you need to:
- Build a systematic churn prevention strategy for a subscription business
- Identify early warning signals that predict customer cancellation
- Design intervention sequences to save at-risk accounts
- Create save offers and retention playbooks for support teams
DO NOT use this skill for win-back campaigns (post-churn), customer acquisition strategies, or pricing restructuring. This is for preventing active customers from leaving.
Core Principle
CHURN PREVENTION STARTS 60 DAYS BEFORE THE CANCELLATION CLICK — BY THE TIME A CUSTOMER ASKS TO CANCEL, YOU HAVE ALREADY LOST MOST OF THE BATTLE.
Phase 1: Brief
Required Inputs
| Input | What to Ask | Default |
|---|
| Product type | "What is your subscription product?" | No default — must be provided |
| Current churn rate | "What is your monthly churn rate?" | Unknown — estimate from available data |
| Billing cycle | "Monthly, annual, or both?" | Monthly |
| Top cancellation reasons | "Why do customers say they leave? Top 3 reasons." | No default — critical input |
| Customer segments | "Do you have different customer tiers or segments?" | Single tier |
| Current retention efforts | "What do you currently do to prevent churn?" | Nothing systematic |
GATE: Confirm the brief before building the playbook.
Phase 2: Plan
Early Warning System
Define 5-8 behavioral signals that predict churn:
## Early Warning Indicators
| Signal | Risk Level | Detection Window |
|--------|-----------|-----------------|
| Login frequency drops 50%+ | High | 14 days |
| Key feature usage stops | High | 7 days |
| Support tickets increase 3x | Medium | 30 days |
| Payment failure (no update) | High | Immediate |
| Team seats decrease | Medium | Billing cycle |
| Downgrade inquiry | High | Immediate |
| NPS score drops below 6 | Medium | Survey cycle |
| No login for 14+ days | Critical | 14 days |
Risk Scoring Model
Create a simple scoring framework:
- Green (0-2 signals): Healthy — standard engagement
- Yellow (3-4 signals): At risk — trigger proactive outreach
- Red (5+ signals): High risk — immediate intervention
GATE: Confirm warning indicators and risk levels before designing interventions.
Phase 3: Write
Intervention Playbook
For each risk level, define the response:
Yellow — Proactive Engagement:
- Automated email: "We noticed you haven't used [feature] recently. Here's a quick tip..."
- In-app message highlighting underused features
- Customer success check-in (if high-value account)
Red — Active Retention:
- Personal email from founder or CS lead
- Offer a 15-minute strategy call
- Share a customer success story relevant to their use case
- Present a save offer (see Save Offer Menu below)
Cancellation Page — Last Resort:
- Exit survey (required, 3 questions max)
- Save offer based on stated reason
- Downgrade option instead of full cancellation
- Pause subscription option (30/60/90 days)
Save Offer Menu
Design offers matched to cancellation reasons:
| Reason | Save Offer | Terms |
|---|
| Too expensive | Discount (20-30% for 3 months) | One-time, non-renewable |
| Not using it enough | Free strategy call + usage guide | Within 7 days |
| Missing a feature | Roadmap preview + beta access | Feature ETA required |
| Switching to competitor | Match competitor pricing or feature | Case-by-case |
| Business closing | Pause subscription | Up to 90 days |
Email Templates
Write 3 core retention emails:
- Check-in email — friendly, value-focused, no mention of churn
- Re-engagement email — highlight a feature they have not tried
- Save email — direct, empathetic, includes a specific offer
Phase 4: Polish
1. Metrics Framework
## Churn Prevention Metrics
- **Monthly churn rate:** Track month-over-month trend
- **Save rate:** % of at-risk accounts retained after intervention
- **Intervention response rate:** % of at-risk users who engage with outreach
- **Save offer acceptance rate:** % who accept a retention offer
- **Time to churn signal:** Average days between first warning signal and cancellation
- **Reactivation rate:** % of paused accounts that resume
2. Team Playbook Card
Create a quick-reference card for support/CS teams:
- Risk signals to watch for
- Approved save offers and their limits
- Escalation path for high-value accounts
- Scripts for common cancellation conversations
3. Quality Checklist
## Churn Prevention Checklist
- [ ] 5-8 early warning indicators defined with detection windows
- [ ] Risk scoring model has clear thresholds (green/yellow/red)
- [ ] Intervention sequence exists for each risk level
- [ ] Save offers are matched to specific cancellation reasons
- [ ] Cancellation page includes exit survey, save offer, and pause option
- [ ] 3 retention email templates are written
- [ ] Metrics framework tracks save rate and intervention effectiveness
- [ ] Support team has a quick-reference playbook card
- [ ] Save offers have clear terms and limits (not unlimited discounts)
Example
Product: Social media scheduling SaaS, $29/month
Churn rate: 8% monthly
Top reasons: Too expensive, not posting enough to justify it, switched to free tool
Check-in email excerpt:
"Hey [Name], I noticed you scheduled 3 posts this month — down from 12 last month. Quick tip: our new batch scheduling feature lets you plan a full week in under 10 minutes. Here is a 2-minute walkthrough. [Watch Video]"
Save offer for "too expensive":
"I get it — every dollar matters when you are running your own business. I can offer you 3 months at $19/month (35% off) while you evaluate the ROI. No strings — if it is still not working after 3 months, cancel with one click."
Anti-Patterns
- Discounting everyone — save offers for price-sensitive churners only. Usage-based churners need value, not discounts.
- Ignoring the warning signals — waiting until cancellation to act is too late. Intervene at yellow, not red.
- Generic "We miss you" emails — reference specific behavior or features. Generic emails feel automated and impersonal.
- No exit survey — if you do not know why people leave, you cannot fix the product. Always collect the reason.
- Unlimited save offers — one discount per customer, with clear terms. Otherwise you train customers to threaten cancellation for discounts.
Recovery
- No churn data exists: Start with an exit survey on the cancellation page. Collect 30 days of data before building the full playbook.
- Churn rate is unknown: Calculate from billing data: (customers lost this month / customers at start of month) x 100.
- Too few customers for patterns: Use qualitative interviews with 5-10 recent churners. Their stories reveal the signals data cannot show yet.
- Team too small for manual outreach: Automate the yellow-tier interventions. Reserve manual effort for red-tier and high-value accounts only.
