Define growth metrics
Skill kimsanguine/signal-to-growth/skills/define-growth-metrics
Evidence-driven skills that turn customer signals into measurable growth.
npx -y skills add kimsanguine/signal-to-growth --skill define-growth-metricsAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 12 days oldThe repository was created 12 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Define SaaS growth metrics as explicit contracts with population, numerator, denominator, window, value event, baseline, target provenance, and counter-metrics. Use when designing activation, retention, revenue, PMF, 성장 지표, KPI, or cohort measurement.
SKILL.md
2.4 KB, 437 tokens by cl100k_base, as published. Nobody here has run it
Define Growth Metrics
Turn a metric name into a reproducible measurement contract. Keep targets tied to a baseline, source, owner, and decision context.
Inputs
Require:
- product and business model;
- user or account entity;
- candidate value event;
- available event and billing data;
- current baseline or an explicit unknown;
- decision owner and review cadence.
Workflow
- State the decision the metric should inform.
- Define the entity and eligible population.
- Define numerator and denominator in plain language.
- Define observation window, cohort maturity, time zone, and exclusions.
- Name the value event and explain why it represents delivered value.
- Record source tables or events and query version.
- Record baseline with date and source, or set it to
null. - Propose a target only with provenance, rationale, and human owner.
- Add at least one counter-metric for quality, risk, or cost.
- Separate activation, retention, revenue, and satisfaction rather than merging them into one score.
Boundaries
- Let the model propose metric candidates and diagnostic questions.
- Use deterministic checks for required fields, denominator, cohort maturity, target provenance, and references.
- Require a person to approve the value event, target, and trade-off.
- Do not use an industry benchmark as a default target.
- Do not calculate immature cohorts as failures.
- Do not treat shipping or signup as demonstrated value without a stated rationale.
Outputs
Create:
metrics.jsonlgrowth-loop-map.mdmeasurement-plan.md- optional query stubs that are clearly marked unverified until executed
Read references/output-contract.md before writing them.
Stop conditions
Stop when the value event is undefined, the denominator cannot be reconstructed, the requested target has no source, or the available data cannot answer the question.
Verification
Run:
python3 scripts/stg.py validate-artifacts artifacts/
Report query generation, local query execution, warehouse execution, and reviewed metric output as separate verification states.