Dashboard Design

Dashboard Design

Overview

A good dashboard answers questions. A great dashboard drives decisions. This skill covers dashboard design principles — choosing the right metrics, visualizing data effectively, and building dashboards that teams actually use.

When to Use

• Building a new dashboard for your product
• Redesigning a dashboard that nobody looks at
• Creating executive reports or weekly metric reviews
• Onboarding a team to data-driven decision-making
• Don't use for: ad-hoc analysis (use notebooks or query tools), operational monitoring (use alerting tools, not dashboards)

Instructions

1. Define the Audience and Purpose

Every dashboard needs ONE clear purpose:

• Execution dashboard: For the team making daily decisions. Real-time or daily.
• Health dashboard: For managers tracking trends. Weekly or monthly.
• Strategic dashboard: For executives overseeing goals. Monthly or quarterly.

2. Choose the Right Metrics

• Actionable → If this metric moves, can someone take action?
• Leading → Predicts future outcomes (not lagging/too late)
• Specific → Clear definition, unambiguous
• Few → 5-9 metrics max per dashboard. More = noise

3. Design the Hierarchy

┌─────────────────────────────────────┐
│  NORTH STAR (top, prominent)        │
│  One number that captures value     │
├────────┬────────┬────────┬─────────┤
│ Input  │ Input  │ Input  │ Input   │
│ Metric │ Metric │ Metric │ Metric  │
│ 1      │ 2      │ 3      │ 4       │
├────────┴────────┴────────┴─────────┤
│  Segments / Breakdowns             │
│  (platform, plan, cohort, etc.)    │
└─────────────────────────────────────┘

4. Visualization Best Practices

• Line charts for trends over time (most common, most useful)
• Bar charts for comparisons between groups
• Single number + sparkline for the North Star metric
• Heatmap for density or cohort analysis
• Table for detailed data (use sparingly)
• ❌ Pie charts (hard to compare), gauges (take too much space), 3D charts (distorting)

5. Add Context

Every metric needs:

• Comparison period: Same period last week, month, year
• Target: What are we aiming for?
• Health indicator: 🟢🟡🔴
• Trend arrow: ↑↓ or flat

6. Make it Actionable

• Add annotations for major events (launches, bugs, marketing campaigns)
• Include a "What to watch" section
• Link to deeper analysis for exploration
• Set up alerts for threshold violations

Sample Output

Team Dashboard — Weekly Active Users:

┌─────────────────────────────────────────┐
│  North Star: 42,500 Weekly Active Teams │ 🟢 +5% WoW
│  Target: 50,000 by Q3                    │
├───────────────┬──────────────┬──────────┤
│ Metric        │ This Week    │ Change   │
│ New signups   │ 1,200        │ +8% ↑    │
│ Activation%   │ 68%          │ +3% ↑    │
│ W+1 Retention │ 55%          │ -2% ↓ 🟡 │
│ Paid conv.    │ 12%          │ Flat →   │
├───────────────┴──────────────┴──────────┤
│ ⚠️ Watch: W+1 retention slipping.       │
│ New onboarding experiment launching EOW │
└─────────────────────────────────────────┘

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: uthumany
• Source: uthumany/uthy-legacy-os
• License: MIT
• Homepage: https://uthumany.github.io/uthy-legacy-os/

Install and usage instructions live in the source repository linked above.