Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Monitors customer health, predicts churn risk, and identifies expansion opportunities using weighted scoring models for SaaS customer success
.claude/skills/borghei-customer-success-manager/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | 132% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 85% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 163% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 183% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 216% | 0% |
The agent operates as an expert customer success manager, driving retention and growth through structured onboarding, health monitoring, risk mitigation, expansion identification, and customer advocacy programs.
Before generating, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
ONBOARDING (0-30d) -> ADOPTION (30-90d) -> VALUE REALIZATION (90d+) -> EXPANSION -> ADVOCACYmarkdown# Customer Onboarding: [Customer Name] ## Pre-Kickoff - [ ] Account setup complete - [ ] Key contacts identified - [ ] Success criteria defined - [ ] Implementation timeline agreed - [ ] Resources allocated ## Week 1: Kickoff - [ ] Kickoff meeting conducted - [ ] Goals and milestones confirmed - [ ] Training schedule set - [ ] Communication channels established ## Week 2-4: Implementation - [ ] Technical setup complete - [ ] Data migration (if applicable) - [ ] Integrations configured - [ ] Initial users trained ## Week 4-8: Adoption - [ ] Power users identified - [ ] Workflow adoption started - [ ] Early wins documented - [ ] Feedback collected ## Handoff (Week 8) - [ ] Onboarding review meeting - [ ] Success metrics baseline - [ ] Ongoing cadence established - [ ] Escalation paths clear
HEALTH SCORE = (Product x 40%) + (Relationship x 30%) + (Outcomes x 30%)
PRODUCT (40%)
Login frequency: [0-10]
Feature adoption: [0-10]
Active users vs. licensed: [0-10]
Support tickets (inverse): [0-10]
RELATIONSHIP (30%)
Executive engagement: [0-10]
Meeting attendance: [0-10]
NPS score: [0-10]
Response time: [0-10]
OUTCOMES (30%)
Goals achieved: [0-10]
ROI demonstrated: [0-10]
Business impact: [0-10]
THRESHOLDS
80-100: Healthy (Green) -- maintain cadence, pursue expansion
60-79: Attention (Yellow) -- increase touchpoints, address gaps
0-59: At Risk (Red) -- activate risk playbook immediatelyCustomer: Acme Corp
Product: (9 + 8 + 9 + 9) / 4 = 8.75 -> weighted: 8.75 x 0.40 = 3.50
Relationship: (8 + 7 + 9 + 8) / 4 = 8.00 -> weighted: 8.00 x 0.30 = 2.40
Outcomes: (8 + 9 + 8) / 3 = 8.33 -> weighted: 8.33 x 0.30 = 2.50
Total: (3.50 + 2.40 + 2.50) x 10 = 84 -> GreenLow Engagement:
Low Adoption:
Executive Change:
Competitor Evaluation:
| Signal | Score | Recommended Action | |--------|-------|--------------------| | High adoption (>80% licensed seats active) | +3 | Explore user expansion | | New department expressing interest | +3 | Schedule discovery call | | Feature requests for premium tier | +2 | Position upgrade path | | Executive engagement increasing | +2 | Propose strategic review | | Contract renewal within 90 days | +2 | Bundle expansion into renewal |
markdown# Quarterly Business Review: [Customer Name] ## Partnership Summary - Customer since: [Date] - Current ARR: $[X] - Users: [X] active / [Y] licensed ## Quarter in Review ### Achievements - [Achievement 1 with metric] - [Achievement 2 with metric] ### Metrics | Metric | Target | Actual | Trend | |--------|--------|--------|-------| | [Metric] | [Target] | [Actual] | up/down/flat | ## Value Delivered - Time saved: [X] hours - Cost reduction: $[Y] - Other impact: [Description] ## Next Quarter Goals 1. [Goal 1 with success metric] 2. [Goal 2 with success metric]
bash# Health score calculator python scripts/health_score.py --customer "Customer Name" # QBR generator python scripts/qbr_generator.py --customer "Customer Name" --quarter Q4 # Risk analyzer python scripts/risk_analyzer.py --portfolio customers.csv # Renewal forecaster python scripts/renewal_forecast.py --period Q1
| Problem | Root Cause | Resolution | |---------|-----------|------------| | Health scores not predicting churn | Model weights are stale or too generic | Recalibrate weights quarterly by comparing predicted scores against actual renewal outcomes. Segment scoring by customer tier, lifecycle stage, and use case. | | Onboarding stalls at Week 2-4 | Technical blockers or lack of internal champion | Escalate to implementation team within 48 hours. Schedule a joint troubleshooting call. If champion is absent, request executive sponsor intervention. | | NPS scores dropping across portfolio | Product issues, unresolved support backlog, or relationship decay | Analyze NPS verbatims for common themes. Prioritize red accounts for immediate outreach. Coordinate with Product on systemic issues. | | Expansion conversations rejected | Timing misaligned with customer value realization | Only initiate expansion after demonstrating measurable ROI. Lead with value recap before any commercial discussion. Wait until health score is Green for 60+ days. | | QBR attendance declining | Content not relevant; too much self-promotion, not enough customer value | Restructure QBR to lead with customer achievements and metrics. Limit product roadmap to items relevant to their use cases. Keep meetings under 45 minutes. | | Executive sponsor changes | Organizational restructuring or M&A activity | Request introduction to new sponsor within 5 business days. Prepare a condensed value summary. Reset success metrics aligned to new sponsor's priorities. | | Customer goes silent (no engagement) | De-prioritization, internal changes, or dissatisfaction not surfaced | Trigger the Low Engagement playbook immediately. Try multiple channels (email, phone, LinkedIn). Engage other known contacts. If no response in 14 days, escalate to your manager for executive outreach. | | Renewal at risk with 60 days remaining | Late identification of churn signals; health score reviewed too infrequently | Increase monitoring cadence to weekly for all renewals within 90 days. Run churn risk scoring monthly. Pre-negotiate renewal terms 120 days before expiry. |
| Metric | Target | Measurement Method | |--------|--------|--------------------| | Gross revenue retention (GRR) | 90%+ | Renewed ARR / Expiring ARR (excluding expansion) | | Net revenue retention (NRR) | 110%+ | (Renewed + Expansion - Contraction) / Beginning ARR | | Logo retention rate | 90%+ | Renewed customers / Total customers up for renewal | | Customer health score accuracy | 80%+ predictive | Percentage of Green accounts that actually renewed | | Time-to-value | Under 30 days | Days from contract signature to first measurable outcome | | QBR completion rate | 100% for accounts above ARR threshold | QBRs delivered / QBRs due per quarter | | NPS score | 50+ | Portfolio-wide NPS from quarterly surveys | | Expansion revenue | 20%+ of book | Expansion ARR / Total managed ARR | | Support escalation resolution | Under 48 hours | Average time from escalation to resolution |
In Scope:
Out of Scope:
Limitations:
| Integration | Direction | Purpose | Handoff Artifact | |-------------|-----------|---------|-----------------| | Account Executive | AE -> CSM | Post-sale handoff with deal context and success criteria | Handoff template with stakeholder map, success criteria, implementation timeline | | Sales Engineer | SE -> CSM | Technical context from pre-sale evaluation | Technical discovery notes, POC results, integration requirements | | Sales Operations | Bidirectional | Renewal forecasting, expansion pipeline tracking, churn reporting | Renewal forecast submissions, health score data exports | | Product Team | CSM -> Product | Feature requests, usage feedback, product issues | Aggregated feedback reports, feature request rankings, bug reports | | Support Team | Support -> CSM | Escalation routing, ticket trends, resolution tracking | Escalation alerts, monthly ticket summaries by account | | Marketing | CSM -> Marketing | Customer stories, references, advocacy program | Case study candidates, reference availability, NPS promoters list | | Finance | Bidirectional | Renewal pricing, credit requests, revenue forecasting | Renewal quotes, churn impact reports, expansion revenue tracking |
Workflow Handoff Protocol:
references/onboarding.md -- Onboarding playbookreferences/health_scoring.md -- Health score methodologyreferences/retention.md -- Retention strategiesreferences/expansion.md -- Expansion playbook| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | fail→pass | 12,241 | 5,916 | -52% | 1 | 1 | 0% | 1,841 | 4,267 | +132% | 0 | 0 | — |
case-01 | fail→fail | 13,485 | 17,402 | +29% | 1 | 1 | 0% | 2,503 | 6,445 | +157% | 0 | 0 | — |
case-02 | fail→fail | 13,075 | 13,168 | +1% | 1 | 1 | 0% | 2,227 | 5,482 | +146% | 0 | 0 | — |
case-03 | fail→fail | 16,948 | 17,198 | +1% | 1 | 1 | 0% | 2,684 | 6,323 | +136% | 0 | 0 | — |
case-04 | fail→fail | 16,640 | 21,980 | +32% | 1 | 1 | 0% | 2,634 | 7,458 | +183% | 0 | 0 | — |
case-05 | fail→fail | 16,767 | 17,351 | +3% | 1 | 1 | 0% | 3,354 | 6,890 | +105% | 0 | 0 | — |
case-06 | fail→pass | 13,010 | 6,579 | -49% | 1 | 1 | 0% | 2,374 | 4,388 | +85% | 0 | 0 | — |
case-07 | fail→pass | 12,950 | 13,951 | +8% | 1 | 1 | 0% | 2,179 | 5,732 | +163% | 0 | 0 | — |
case-08 | pass→pass | 8,800 | 5,549 | -37% | 1 | 1 | 0% | 1,809 | 4,585 | +153% | 0 | 0 | — |
case-09 | fail→pass | 11,952 | 11,561 | -3% | 1 | 1 | 0% | 1,913 | 5,408 | +183% | 0 | 0 | — |
case-10 | fail→pass | 10,201 | 9,722 | -5% | 1 | 1 | 0% | 1,602 | 5,064 | +216% | 0 | 0 | — |
case-11 | fail→pass | 7,546 | 4,080 | -46% | 1 | 1 | 0% | 1,174 | 4,068 | +247% | 0 | 0 | — |
case-12 | fail→pass | 10,704 | 5,163 | -52% | 1 | 1 | 0% | 1,796 | 4,343 | +142% | 0 | 0 | — |
case-13 | fail→pass | 13,170 | 16,103 | +22% | 1 | 1 | 0% | 2,072 | 5,832 | +181% | 0 | 0 | — |
case-14 | fail→pass | 7,942 | 3,818 | -52% | 1 | 1 | 0% | 1,300 | 4,016 | +209% | 0 | 0 | — |
case-15 | fail→pass | 7,648 | 5,111 | -33% | 1 | 1 | 0% | 1,345 | 4,257 | +217% | 0 | 0 | — |
case-16 | fail→pass | 4,733 | 1,883 | -60% | 1 | 1 | 0% | 659 | 3,678 | +458% | 0 | 0 | — |
case-17 | fail→pass | 13,242 | 13,743 | +4% | 1 | 1 | 0% | 2,076 | 5,692 | +174% | 0 | 0 | — |
case-19 | pass→pass | 9,958 | 6,802 | -32% | 1 | 1 | 0% | 1,918 | 4,572 | +138% | 0 | 0 | — |
case-20 | fail→fail | 10,108 | 9,363 | -7% | 1 | 1 | 0% | 1,453 | 4,862 | +235% | 0 | 0 | — |
case-21 | fail→pass | 6,829 | 2,537 | -63% | 1 | 1 | 0% | 1,183 | 3,797 | +221% | 0 | 0 | — |
case-22 | pass→pass | 10,239 | 3,880 | -62% | 1 | 1 | 0% | 1,650 | 4,107 | +149% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.