multiAI Summary Pending
Churn Risk Analyzer
Identify customers most likely to churn before they leave. Uses behavioral signals, usage patterns, and engagement data to score accounts and recommend retention actions.
3,556 stars
byopenclaw
Installation
Claude Code / Cursor / Codex
$curl -o ~/.claude/skills/afrexai-churn-analyzer/SKILL.md --create-dirs "https://raw.githubusercontent.com/openclaw/skills/main/skills/1kalin/afrexai-churn-analyzer/SKILL.md"
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/afrexai-churn-analyzer/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How Churn Risk Analyzer Compares
| Feature / Agent | Churn Risk Analyzer | Standard Approach |
|---|---|---|
| Platform Support | multi | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Identify customers most likely to churn before they leave. Uses behavioral signals, usage patterns, and engagement data to score accounts and recommend retention actions.
Which AI agents support this skill?
This skill is compatible with multi.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Churn Risk Analyzer Identify customers most likely to churn before they leave. Uses behavioral signals, usage patterns, and engagement data to score accounts and recommend retention actions. ## When to Use - Customer success reviews - Quarterly retention planning - When usage data or support ticket logs are available - Proactive outreach prioritization ## How It Works ### 1. Gather Data Ask the user for available data sources: - **Usage metrics** (logins, feature adoption, API calls) - **Support tickets** (frequency, sentiment, resolution time) - **Billing history** (downgrades, late payments, discount requests) - **Engagement signals** (email opens, meeting attendance, NPS scores) If no structured data, work from what the user describes qualitatively. ### 2. Score Each Account Apply this risk framework: | Signal | Weight | High Risk Indicator | |--------|--------|-------------------| | Usage decline (30d) | 25% | >30% drop | | Support ticket spike | 20% | 2x+ above baseline | | Champion departure | 20% | Key contact left | | Contract timing | 15% | <90 days to renewal | | Payment behavior | 10% | Late/disputed invoices | | Engagement drop | 10% | No response to last 3 outreach | **Score**: 0-100 (higher = more likely to churn) ### 3. Categorize - **🔴 Critical (75-100)**: Immediate intervention needed - **🟡 At Risk (50-74)**: Schedule check-in this week - **🟢 Healthy (25-49)**: Monitor monthly - **💚 Thriving (0-24)**: Expansion opportunity ### 4. Recommend Actions For each at-risk account, suggest specific retention plays: - Executive sponsor call - Custom success plan - Feature training session - Pricing review / loyalty offer - Roadmap preview (show upcoming value) ### 5. Output Generate a retention report with: - Ranked list of accounts by churn risk - Top 3 actions per critical account - Estimated revenue at risk - 30/60/90 day retention calendar ## Integration Notes - Works with CSV exports, CRM data, or manual input - Pair with the [AfrexAI Context Packs](https://afrexai-cto.github.io/context-packs/) for industry-specific retention strategies ($47/pack) - For a full AI-powered retention system built for your business, try the [AI Revenue Calculator](https://afrexai-cto.github.io/ai-revenue-calculator/) to see what you're leaving on the table