이탈 위험 탐지기 - 떠나기 전에 위험 계정를 찾습니다 — Claude Skill

이탈 위험 탐지기

너무 늦기 전에 churn risk가 있는 account를 드러냅니다. support, communication, usage pattern의 signal을 모아 specific save action이 포함된 scored risk report로 만듭니다.

대상: CS platform이 없는 early-stage team(Gainsight, ChurnZero 없음). customer spreadsheet, Slack channel, support inbox가 있는 상태입니다. 이 스킬은 그 raw signal을 actionable churn risk list로 바꿉니다.

언제 사용할지

• "어떤 customer가 churn risk가 있나요?"
• "weekly churn risk scan을 실행해줘"
• "걱정해야 할 account를 flag해줘"
• "한동안 연락 없는 사람은 누구인가요?"
• "customer health report를 만들어줘"

Phase 0: Intake

Account data

1. Customer list — CSV 또는 sheet: company name, primary contact email, contract value(MRR/ARR), contract start date, renewal date(알고 있다면)
2. Product/service type — 무엇에 비용을 지불하나요? (expected engagement calibration에 도움)

Signal sources(가진 것을 제공)

3. Support tickets — Intercom, Zendesk, email export(CSV: customer, date, subject, status, resolution time)
4. Slack channel history — customer Slack channel 또는 shared channel message
5. NPS/CSAT scores — score와 comment가 있는 최근 survey result
6. Usage data — tracking하는 metric: login, API call, feature used, active user(CSV export)
7. Email/communication log — account별 last touchpoint(date + context)
8. Billing data — payment failure, downgrade, discount request

Calibration

9. "healthy"는 어떤 모습인가요? — healthy customer를 설명합니다(예: "매일 login, 3+ feature 사용, email에 24시간 안에 response")
10. Known churn reasons — 과거 customer가 왜 churn했나요? (signal weighting에 도움)

Phase 1: Signal extraction

1A: Support signal analysis

Support ticket data에서 account별로 계산합니다:

1B: Communication signal analysis

Slack/email history에서:

1C: Usage signal analysis(data가 있으면)

1D: Commercial signal analysis

Phase 2: Risk scoring

Scoring model

각 account는 composite risk score(0-100)를 받습니다:

Risk Score = Σ (signal_weight × signal_present)

Weights:
  Critical signal = 25 points each
  High signal     = 15 points each
  Medium signal   = 8 points each
  Low signal      = 3 points each

Score cap: 100

Risk tiers

Phase 3: Save play generation

각 Red와 Orange account에 specific save play를 생성합니다:

Save play template

ACCOUNT: [Company Name]
RISK TIER: [Red/Orange]
RISK SCORE: [X/100]
MRR/ARR: $[X]

SIGNALS DETECTED:
- [Signal 1] — [Evidence: specific data point]
- [Signal 2] — [Evidence]
- [Signal 3] — [Evidence]

ROOT CAUSE HYPOTHESIS:
[1-2 sentences: What do you think is actually going wrong?
 E.g., "Champion left the company and new stakeholder hasn't been onboarded"
 or "They hit a technical limitation with [feature] that's blocking their primary use case"]

RECOMMENDED SAVE PLAY:
1. [Immediate action — e.g., "Schedule a call with [contact] this week"]
2. [Follow-up — e.g., "Send a personalized Loom showing how to solve [specific issue]"]
3. [Structural fix — e.g., "Assign a dedicated onboarding session for new stakeholder"]

TALK TRACK:
"[2-3 sentences the CSM/founder can use to open the conversation naturally,
 without saying 'we noticed you might be churning']"

ESCALATION TRIGGER:
If [specific condition] by [date], escalate to [founder/CEO call].

Phase 4: 출력 형식

# Churn Risk Report — Week of [DATE]
Total accounts scanned: [N]
Data sources: [list what was available]

---

## Risk Summary

| Tier | Count | Total MRR at Risk |
|------|-------|-------------------|
| 🔴 Red (Critical) | [N] | $[X] |
| 🟠 Orange (Elevated) | [N] | $[X] |
| 🟡 Yellow (Early Warning) | [N] | $[X] |
| 🟢 Green (Healthy) | [N] | $[X] |

**Total MRR at risk (Red + Orange):** $[X] ([Y]% of total MRR)

---

## 🔴 Critical Risk Accounts

### [Company Name 1] — Score: [X]/100 | MRR: $[X]
**Signals:** [bullet list]
**Root cause:** [hypothesis]
**Save play:** [specific actions]
**Owner:** [who should act]
**Deadline:** [date]

### [Company Name 2] — ...

---

## 🟠 Elevated Risk Accounts

### [Company Name] — Score: [X]/100 | MRR: $[X]
**Signals:** [bullet list]
**Recommended action:** [1-2 sentences]

---

## 🟡 Early Warning Accounts

| Account | Score | Key Signal | Suggested Action |
|---------|-------|------------|-----------------|
| [Name] | [X] | [Signal] | [Action] |
| [Name] | [X] | [Signal] | [Action] |

---

## Trends vs Last Week

- Accounts moved Red → Green: [list — wins!]
- Accounts moved Green → Yellow/Orange: [list — new risks]
- Accounts churned since last report: [list]

---

## Signal Distribution

| Signal Type | Accounts Affected |
|------------|-------------------|
| Support ticket spike | [N] |
| Gone silent | [N] |
| Usage decline | [N] |
| Competitor mention | [N] |
| Payment issue | [N] |
| Champion disengagement | [N] |

---

## Recommended Focus This Week

1. **[Account]** — [Why + what to do]
2. **[Account]** — [Why + what to do]
3. **[Account]** — [Why + what to do]

clients/<client-name>/customer-success/churn-risk/risk-report-[YYYY-MM-DD].md에 저장합니다.

Scheduling

매주 실행:

0 8 * * 1 python3 run_skill.py churn-risk-detector --client <client-name>

비용

필요한 도구

• CSV/sheet의 input data(support ticket, usage, NPS)
• Optional: communication signal을 위한 Slack channel reading
• 외부 API 비용 없음. 순수 analysis

트리거 문구

• "어떤 customer가 risk가 있나요?"
• "churn risk scan을 실행해줘"
• "weekly customer health report"
• "at-risk account를 flag해줘"