mcpField
U
usage-health-scorecard

usage-health-scorecard

github.com/gtmagentsData & databases42864 installsUpdated 30w ago
Repository

About usage-health-scorecard

A skill for product, growth, and customer success teams tracking usage health across customer cohorts. It groups metrics, segments results, sets thresholds, and records alerts and follow-up actions.

Key features

  1. 01Activation and engagement metrics
  2. 02Persona and cohort segmentation
  3. 03Traffic-light thresholds
  4. 04Drop and anomaly alerts
  5. 05Remediation action tracking

Use cases

  1. 01Track activation and retention by cohort
  2. 02Assess product health during PLG experiments
  3. 03Share health snapshots with team leads

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/gtmagents/gtm-agents --skill usage-health-scorecard

After installing, type/usage-health-scorecardin Claude Code to confirm it loaded.

Details

From the project README

Usage Health Scorecard Skill

When to Use

  • Tracking activation/retention metrics for self-serve and hybrid customer cohorts.
  • Monitoring the impact of PLG experiments on core product health.
  • Sharing health snapshots with product, growth, and CS leads.

Framework

  1. Metric Groups – activation, engagement, collaboration, monetization, support load.
  2. Segmentation – persona, plan, cohort, region, product area.
  3. Thresholds – traffic light ranges for each metric with warning/critical bands.
  4. Alerting – notification rules for drops, anomalies, or experiment impacts.
  5. Action Registry – log remediation steps, owners, due dates, and results.

Templates

  • Scorecard layout (metric, current, target, delta, owner).
  • Dashboard wireframe with spark lines + annotations.
  • Action tracker for follow-up tasks linked to health changes.

Tips

  • Pair with launch-in-app-experiments to watch guardrails post-launch.
  • Include qualitative signals (support tags, NPS) to contextualize telemetry.
  • Archive snapshots to tell longitudinal stories in QBRs.

Comments

Markdown supported