L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

churn-analyzer

churn-analyzer

Analyze customer churn patterns and create retention strategies. TRIGGERS - Use when user wants to reduce churn, improve retention, or understand why customers leave.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: churn-analyzer description: | Analyze customer churn patterns and create retention strategies. TRIGGERS - Use when user wants to reduce churn, improve retention, or understand why customers leave.

Churn Analyzer

Overview

Analyzes churn patterns, identifies root causes, and creates actionable retention strategies.

Workflow

Step 1: Gather Context

  1. Business model: SaaS, service, product?
  2. Current churn rate: Monthly/annual?
  3. Customer segments: Different tiers or types?
  4. Known reasons: Why do customers say they leave?
  5. Available data: What customer data do you have?

Step 2: Analyze Churn

Churn calculation:

Monthly churn = Customers lost / Starting customers × 100
Annual churn = 1 - (1 - monthly churn)^12
Revenue churn = MRR lost / Starting MRR × 100

Analysis dimensions:

  • By segment (plan tier, industry, size)
  • By tenure (when do they leave?)
  • By engagement (usage before churn)
  • By acquisition source (where they came from)
  • By season (time patterns)

Output Format

# Churn Analysis: [Company]

## Current State
- **Monthly churn**: [X%]
- **Annual churn**: [X%]
- **Revenue impact**: $[X] lost/month
- **Industry benchmark**: [X%]

## Churn Patterns

### By Tenure
| Period | Churn Rate | Insight |
|--------|-----------|---------|
| Month 1 | X% | [early churn = onboarding problem] |
| Month 2-3 | X% | [value realization gap] |
| Month 6+ | X% | [engagement decay] |

### By Segment
[Breakdown by customer type]

## Root Cause Analysis
1. **[Cause 1]** — [% of churn] — [evidence]
2. **[Cause 2]** — [% of churn] — [evidence]
3. **[Cause 3]** — [% of churn] — [evidence]

## Retention Strategy

### Quick Wins (This Month)
1. [Action]: Expected impact [X% reduction]
2. [Action]: Expected impact [X% reduction]

### Medium-Term (This Quarter)
1. [Action]: Expected impact [X% reduction]

### Long-Term (This Year)
1. [Action]: Expected impact [X% reduction]

## Financial Impact
If churn reduces from [X%] to [Y%]:
- Monthly revenue saved: $[X]
- Annual impact: $[X]
- LTV improvement: [X%]

Quality Checklist

  • Churn calculated correctly (logo + revenue)
  • Patterns identified by multiple dimensions
  • Root causes ranked by impact
  • Retention strategies are specific and actionable
  • Financial impact quantified
Skills enregistrées