metric-dashboard

Metric Dashboard Skill

Design a comprehensive metric dashboard and KPI tracking plan for any product or feature.

When to Use

• User needs to define metrics for a new product or feature
• User is setting up monitoring and alerting
• User needs to design a dashboard layout
• User says /metric-dashboard followed by the product/feature
• Any time measurement strategy needs to be defined

Framework: Metric Dashboard Design (5 Steps)

Step 1: Define the Metric Hierarchy

North Star Metric (NSM): The single metric that best captures the value your product delivers.

• Must reflect user value, not just business value
• Must be measurable with current instrumentation
• Formula: NSM = [engagement unit] per [user segment] per [time period]

Decompose into a metric tree:

North Star Metric
├── Input Metric A (e.g., new users)
│   ├── Sub-metric A1
│   └── Sub-metric A2
├── Input Metric B (e.g., activation rate)
│   ├── Sub-metric B1
│   └── Sub-metric B2
└── Input Metric C (e.g., retention)
    ├── Sub-metric C1
    └── Sub-metric C2

Step 2: Categorize Metrics

Product Metrics:

• Acquisition: How users find you (sign-ups, installs, registrations)
• Activation: First value moment (onboarding completion, first action)
• Engagement: Core usage (DAU/MAU, session length, feature adoption)
• Retention: Coming back (D1/D7/D30, cohort retention curves)
• Revenue: Monetization (ARPU, conversion, LTV, churn)

Technical Metrics:

• Performance: Latency (p50, p95, p99), throughput, error rate
• Reliability: Uptime, incident count, MTTR
• Infrastructure: CPU/memory utilization, cost per request

AI/ML Metrics (if applicable):

• Quality: Accuracy, hallucination rate, eval scores
• Safety: Content policy violation rate, false refusal rate
• Cost: Cost per inference, token usage
• Latency: Time to first token, tokens per second

Business Metrics:

• Revenue: MRR, ARR, revenue growth rate
• Unit economics: CAC, LTV, LTV/CAC ratio
• Market: Market share, competitive win rate