mcpField
R
retention-dashboard

retention-dashboard

github.com/gtmagentsData & databases42865 installsUpdated 30w ago
Repository

About retention-dashboard

retention-dashboard is a retention analytics toolkit skill for customer success, lifecycle marketing, and executive teams. It provides templates for defining metrics and segments, and planning dashboards, alerts, and annotation workflows.

Key features

  1. 01Define retention and health metrics
  2. 02Segment by plan, region, and other dimensions
  3. 03Plan dashboard and visualization layouts
  4. 04Set threshold-based alerts
  5. 05Record commentary, actions, and owners

Use cases

  1. 01Automate recurring retention reviews
  2. 02Track pilot and save-play outcomes
  3. 03Give segment owners drillable dashboards

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/gtmagents/gtm-agents --skill retention-dashboard

After installing, type/retention-dashboardin Claude Code to confirm it loaded.

Details

From the project README

Retention Dashboard Toolkit Skill

When to Use

  • Automating retention reviews for CS, lifecycle marketing, or execs.
  • Tracking pilot outcomes for adoption/save plays.
  • Providing drillable dashboards to segment owners.

Framework

  1. Metric Definitions – ARR retention, logo retention, expansion %, health scores.
  2. Cohort Dimensions – plan, persona, region, industry, product, acquisition channel.
  3. Visualization Layout – summary tiles, cohort heatmaps, waterfall, signal callouts.
  4. Alerting Layer – thresholds for Slack/email alerts when metrics breach targets.
  5. Annotation Workflow – capture commentary, actions, and follow-up owners.

Templates

  • BI dashboard spec (metrics, dimensions, filters, refresh cadence).
  • Weekly retention digest format.
  • Alert template with context + call to action.

Tips

  • Normalize metrics (e.g., ARR, accounts, seats) to avoid confusion.
  • Tie charts to plays so stakeholders know what to do next.
  • Pair with activation-map to log actions triggered by signals.

Comments

Markdown supported