dashboard-critique

Use when reviewing a dashboard, KPI page, or reporting artifact and the AI should assess metric clarity, structure, comparability, and decision usefulness.

Dashboard Critique

Purpose

Assess whether a dashboard helps people make better decisions or merely displays numbers.

When to use

Use this skill when:

  • reviewing a KPI dashboard
  • improving an existing report
  • deciding whether a dashboard is fit for stakeholder use

When not to use

Do not use this skill when:

  • the task is to diagnose a metric change in detail
  • there is no reporting artifact to review

Required thinking discipline

  • Judge the dashboard against decisions, not aesthetics alone.
  • Check metric definitions, baselines, and comparability.
  • Look for ways the dashboard could mislead a stakeholder.
  • Evidence constraint: Every conclusion must cite specific data — a number, a rate, a segment, or a timeframe. Do not speculate without evidential basis. If data is insufficient, state what is missing rather than asserting an unsupported inference.

Workflow

  1. Identify the audience and decision use case.
  2. Review metric labels and definitions.
  3. Review baselines, comparisons, and time context.
  4. Review segmentation and drill-down usefulness.
  5. Identify likely misreads or missing context.
  6. Recommend the smallest changes that materially improve decision value.

Output format

  • Intended audience
  • What works
  • What is unclear or risky
  • Missing context
  • Recommended changes
  • Priority order

Good example

The dashboard shows weekly active users but not the baseline, target, or segmentation. A manager can see the count moved, but not whether the movement is meaningful or where it came from.

Bad example

The dashboard looks busy and should be cleaner.

Why this is bad:

  • it is mostly aesthetic
  • it ignores business usefulness
  • it does not mention metric clarity

Practical notes

  • A good dashboard usually answers a recurring decision question.
  • If a metric can be interpreted multiple ways, the dashboard should not force the audience to guess.

Optional variants

  • Executive dashboards: focus on signal hierarchy and decision framing.
  • Analyst dashboards: focus on drill paths, definitions, and segment cuts.