churn-analysis
When the user needs to identify at-risk accounts, understand why customers are leaving, reduce churn rate, build health scores, design save plays, or create win-back campaigns.
74
68%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/churn-analysis/SKILL.mdQuality
Discovery
72%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description has strong trigger terms and clear distinctiveness in the customer churn/retention domain, but it lacks an explicit 'what this skill does' statement. It reads entirely as a 'when to use' clause without first establishing the skill's core capabilities or outputs. Adding a concrete capability statement before the trigger conditions would significantly improve it.
Suggestions
Add a leading 'what' statement describing the skill's concrete outputs, e.g., 'Analyzes customer data to predict churn, generates retention strategies, and builds customer health scoring models.'
Restructure to separate capabilities from triggers: start with specific actions/outputs, then follow with the existing 'Use when...' clause.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (customer churn/retention) and lists several actions like 'identify at-risk accounts', 'build health scores', 'design save plays', and 'create win-back campaigns'. However, these are more like task categories than concrete specific actions—it doesn't describe what the skill actually produces or how it operates. | 2 / 3 |
Completeness | The description provides a 'when' clause ('When the user needs to...') but lacks a clear 'what does this do' statement. It only describes trigger conditions without explaining what the skill actually does or produces. The 'what' is only implied through the trigger terms. | 2 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'at-risk accounts', 'customers are leaving', 'churn rate', 'health scores', 'save plays', 'win-back campaigns'. These cover a good range of terms a user working on customer retention would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | The description carves out a clear niche around customer churn analysis and retention strategies. Terms like 'churn rate', 'health scores', 'save plays', and 'win-back campaigns' are highly specific to this domain and unlikely to conflict with other skills. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a thorough, highly actionable churn analysis skill with excellent frameworks (risk scoring model, churn driver taxonomy, cancel flow design) and strong examples showing exact output formats. Its main weaknesses are verbosity — several sections include educational context and industry benchmarks that pad the token count — and the lack of validation checkpoints in the workflow. The inline detail level is appropriate for a standalone reference but would benefit from splitting into sub-files for progressive disclosure.
Suggestions
Add explicit validation checkpoints to the workflow, e.g., 'Verify data completeness before scoring — flag accounts with insufficient signals for manual review' and 'Cross-check top-10 risk scores against CS team intuition before distributing the scorecard.'
Move detailed framework sections (Signal Extraction Categories, Cancel Flow Design, Dunning, Win-Back) into separate referenced files to reduce the main skill's token footprint and improve progressive disclosure.
Trim educational/statistical content that Claude doesn't need to execute the skill, such as industry benchmark percentages ('saves 10-20%', 'recovers 30-50%') and platform name-drops ('Gainsight or ChurnZero').
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly comprehensive but includes some unnecessary verbosity. Phrases like 'Especially relevant for seed/Series A teams managing customers manually without dedicated CS platforms like Gainsight or ChurnZero' and general industry statistics (e.g., 'saves 10-20% of users') add context Claude doesn't need. The frameworks section is detailed but could be tighter — some explanations are more educational than instructional. | 2 / 3 |
Actionability | The skill provides highly concrete, actionable guidance: specific scoring models with point values, risk tier thresholds with response timelines, a structured cancel flow with numbered steps, dunning retry schedules, and win-back timing windows. The examples show exact output formats with tables and save play briefs that Claude can directly replicate. | 3 / 3 |
Workflow Clarity | The 6-step workflow is clearly sequenced and logical, but it lacks explicit validation checkpoints. There's no verification step to confirm data quality before scoring, no checkpoint to validate risk scores against known outcomes, and no feedback loop for iterating on the scoring model. For a process that drives customer interventions, validation gaps are notable. | 2 / 3 |
Progressive Disclosure | The skill has good section structure with clear headers, but it's quite long (~200+ lines) with detailed frameworks (signal categories, scoring model, cancel flow design, dunning, win-back) all inline. The signal extraction categories, cancel flow design, and dunning sections could be split into referenced files. The 'Related Skills' section is well-done with clear cross-references. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- shawnpang/startup-founder-skills
- Commit
4ad31b4
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.