Plan a Data Dashboard Layout and Metrics
Plan a full dashboard with curated KPIs, chart types, layout, and vanity-metric warnings.
This prompt acts as a data-viz consultant that designs your dashboard before you build it — choosing which KPIs to show, how to group and chart them, and crucially which 'metrics' to throw out as vanity. It's for the planning stage, when it's far cheaper to argue about whether 'total pageviews' belongs on the screen than to rebuild a Looker dashboard later. The vanity-metric flag is the standout step: it stops you from shipping a dashboard that looks impressive but drives no decisions.
You are a data visualization expert. Help me plan a dashboard for [AUDIENCE: e.g. executives / ops team / marketing] tracking [BUSINESS DOMAIN]. (1) Recommend 6-10 KPIs to display with the rationale for each. (2) Group them into logical sections. (3) Recommend chart type for each metric (bar, line, gauge, etc.) and why. (4) Flag any metrics that are vanity metrics and suggest better alternatives. (5) Propose a layout in text form (top row, left panel, etc.).
What you can expect back
Recommended KPIs (grouped) GROWTH - MRR - the headline revenue number; line chart to show trend - New MRR vs Churned MRR - net movement; stacked bar by month - New customers - leading indicator; line chart RETENTION - Net revenue retention (NRR) - the single best health metric; gauge vs 100% target - Logo churn rate - customer loss; line chart with target band - Activation rate - % reaching first value; bar by cohort EFFICIENCY - CAC payback (months) - growth sustainability; single big-number tile Vanity metrics flagged: - 'Total signups' -> replace with 'activated accounts', since signups that never activate inflate the number without revenue. - 'Total registered users (all time)' -> replace with 'active accounts (30-day)'. Layout - Top row: MRR + NRR (the two numbers leadership reads first) - Middle: New vs Churned MRR, logo churn trend - Bottom: activation, CAC payback, customer count
Illustrative example — your results will vary by tool and inputs.
Get sharper results
- 01Name the decision the dashboard supports ('should we keep spending on acquisition?') so the model selects metrics that answer a question rather than just available ones.
- 02Tell it your review cadence — a weekly exec dashboard needs trend lines, while a real-time ops dashboard needs current-state gauges and thresholds.
- 03Ask it to specify the comparison or target for each metric (vs last period, vs goal), since a number with no benchmark is undecidable at a glance.
- 04Push back with 'which 3 of these matter most if we could only show 3' to force ruthless prioritization before you build all ten.
- 05Have it note the data source and likely refresh frequency per KPI so you can spot metrics that are expensive or impossible to wire up early.
Adapt it for your case
Constrain it to 'must fit one wall-mounted screen, glanceable from across the room' to get fewer, bigger, status-style tiles.
Swap the domain for a marketing funnel and ask it to organize KPIs by funnel stage (awareness, acquisition, activation, retention).
Add 'I'm building this in Looker Studio / Tableau / Power BI' so it suggests chart types and layout patterns that tool supports well.
Common questions
What makes a metric a 'vanity metric'?
One that looks good and grows over time but doesn't inform a decision or correlate with success — like cumulative signups or total pageviews. The prompt asks the model to swap these for actionable equivalents.
Why limit it to 6-10 KPIs?
A dashboard with 25 numbers is read as zero. Constraining the count forces the genuinely decision-driving metrics to the surface and keeps the layout glanceable.
Can it design the actual queries?
Ask it to — follow up with 'write the SQL or formula for each KPI given my schema' and paste your table structure. The plan and the implementation can come from the same conversation.
You may also need
Analyze an A/B Test for Statistical Significance
Statistically analyze an A/B test result with significance testing and a ship/no-ship recommendation.
Choose the Right Chart Type for Your Data
Get a chart type recommendation with rationale, alternatives, and common pitfalls for your data.
Write a SQL Query From a Business Question
Translate a business question into a clean, commented SQL query against your schema.
Analyze a Dataset With Pandas Step-by-Step
Generate step-by-step Pandas EDA code covering nulls, outliers, and a business question.