Skip to content

hoavdc/CodexKit / skills/codexkit-churn-risk-analyzer

codexkit-churn-risk-analyzer

Unverified

Build customer health scores and churn risk models using leading indicators. Define health dimensions (Usage, Engagement, Support, Sentiment, Contract), weight and score customers, segment into Healthy / At-Risk / Red, and generate intervention playbooks. Use during QBR prep, customer success reviews, or when churn spikes.

Churn Risk Analyzer

When to Use

  • During monthly or quarterly customer health reviews
  • When churn rate increases and root causes are unclear
  • When building a customer health scoring model for the first time
  • When preparing retention plays for at-risk accounts

Procedure

Step 1 — Define Health Dimensions

DimensionWeightIndicators
Usage30%Login frequency, feature adoption, DAU/MAU ratio
Engagement25%Support ticket sentiment, NPS, event attendance
Support15%Ticket volume, escalations, resolution time
Sentiment15%NPS score, CSAT, qualitative feedback
Contract15%Time to renewal, expansion signals, payment health

Adjust weights by business model (self-serve vs enterprise).

Step 2 — Score Each Dimension

ScoreLevelCriteria
5HealthyStrong usage, positive sentiment, expanding
4GoodRegular usage, neutral/positive feedback
3ModerateDeclining trends, some concerns
2At-RiskSignificant decline, negative signals
1CriticalMinimal engagement, escalations, churn signals

Step 3 — Calculate Health Score

Health Score = Σ (Dimension Score × Weight)

Step 4 — Segment Customers

Health ScoreSegmentAction
4.0–5.0🟢 HealthyExpansion play, referral ask
2.5–3.9🟡 At-RiskProactive outreach, value reinforcement
1.0–2.4🔴 RedExecutive sponsor call, save plan

Step 5 — Intervention Playbooks

🟢 Healthy:

  • Identify expansion opportunities (upsell, cross-sell)
  • Request referral or case study
  • Invite to advisory board or beta programs

🟡 At-Risk:

  • Schedule CSM check-in within 48 hours
  • Re-onboard on underused features
  • Share success stories from similar companies
  • Offer training session or office hours

🔴 Red:

  • Executive sponsor call within 24 hours
  • Create 30-day save plan with specific milestones
  • Offer concessions if justified (credit, extended trial)
  • Prepare for graceful offboarding if save fails

Inputs

InputRequiredFormat
Customer listYesAccount names with contract data
Usage dataYesLogin counts, feature adoption metrics
Support dataRecommendedTicket count, CSAT, escalations
NPS/sentiment dataRecommendedScores or qualitative feedback
Contract detailsRecommendedRenewal dates, ARR, payment status

Output

## Churn Risk Report — [Period]

### Portfolio Health Summary

| Segment | Count | % of Base | ARR at Risk |
|---------|-------|-----------|-------------|
| 🟢 Healthy | 120 | 60% | — |
| 🟡 At-Risk | 55 | 27.5% | $820K |
| 🔴 Red | 25 | 12.5% | $450K |

### Top 10 At-Risk Accounts

| Account | Health Score | Top Risk Factor | ARR | Renewal | CSM Action |
|---------|-------------|-----------------|-----|---------|------------|
| Acme Corp | 2.8 | Usage ↓ 40% | $120K | 60 days | Re-onboarding |
| Beta Inc | 2.5 | NPS dropped to 4 | $85K | 90 days | Exec call |
| [etc.] | | | | | |

### Intervention Queue

| Priority | Account | Action | Owner | Deadline |
|----------|---------|--------|-------|----------|
| 1 | Acme Corp | Schedule exec sponsor call | VP CS | This week |
| 2 | Beta Inc | Feature re-onboarding | CSM | Next week |

### Churn Risk Drivers (Pareto)
1. Usage decline (40% of at-risk accounts)
2. Support escalation unresolved (25%)
3. Champion left the company (20%)
4. Contract/pricing dissatisfaction (15%)

Definition of Done

  • Health dimensions defined with weights
  • Each customer scored per dimension
  • Weighted health score calculated
  • Customers segmented into 🟢/🟡/🔴
  • Intervention playbook per segment
  • Top at-risk accounts listed with actions and owners
  • Churn risk drivers identified (Pareto)

Examples

Prompt

We have 200 B2B customers. Churn rate spiked from 5% to 8% this quarter.
Here is our customer data: [paste usage, support, NPS data]
Build a churn risk model with health scores, segment our portfolio,
and create intervention playbooks for at-risk accounts.

Quality Criteria

  • Data sources and assumptions are explicitly stated
  • Calculations are reproducible from provided inputs
  • Visualizations or tables have clear labels, units, and time ranges
  • Caveats and confidence levels are documented for estimates

Verification (4C)

CheckQuestion
CorrectnessAre formulas, aggregations, and statistical methods applied correctly?
CompletenessDoes the analysis cover all requested metrics and time ranges?
Context-fitAre the chosen metrics relevant to the business question being answered?
ConsequenceIf this data were used for a decision today, what blind spots remain?

Edge Cases

  • Missing or incomplete data — Document gaps and their potential impact on conclusions. Provide ranges instead of point estimates.
  • Outliers skewing results — Report with and without outliers. Document the decision to include or exclude.
  • Changing data definitions mid-period — Split analysis at the change boundary and note the schema difference.

Changelog

  • v1.0.0 — Initial release
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 318kUpdated Claude CodeCodexUnverifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 318kUpdated Claude CodeCodexUnverifiedInstall