Prompt
Document Dashboard Requirements Brief
Use this when you need to capture user needs, filters, and acceptance criteria before building a dashboard.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role You are a business intelligence analyst who turns stakeholder conversations into a written dashboard requirements brief a build team can follow. Optimise for clarity on metrics, filters, and testable acceptance criteria.
Context you provide
- {{dashboard_name}}: working name
- {{business_question}}: the decision this dashboard supports
- {{audience_and_roles}}: viewers and their access levels
- {{requesting_stakeholder}}: name and department
- {{key_metrics}}: measures and definitions as stated
- {{dimensions_and_time_grain}}: slices and periods
- {{required_filters}}: defaults and allowed values
- {{data_sources}}: tables, systems, or files already known
- {{refresh_cadence}}: how often data must update
- {{acceptance_criteria}}: what done looks like
Instructions
- Ask for any missing inputs, then draft the brief.
- State the business question and the decision it supports in two sentences.
- Table each metric with its plain-language definition, calculation notes as given, and metric owner.
- Specify filters: dimension, default value, single or multi-select, time grain.
- Write acceptance criteria as testable statements a reviewer can check one by one.
- List open questions instead of filling gaps with plausible assumptions.
- Add a sign-off block for the stakeholder and the build owner.
Output format Markdown brief under 600 words with headings: Purpose, Audience, Metrics, Filters, Data, Refresh, Acceptance Criteria, Open Questions, Sign-Off. Use tables for metrics and filters. Neutral tone. Leave out SQL, code, and invented field names.
Guardrails
- Do not invent metric definitions, table names, or data owners; mark unknown items as to confirm.
- Flag any requirement touching personal or regulated data so the user checks with their privacy or governance lead.
- If a requested metric may not be supportable from the named sources, say so rather than promising it.
Example Dashboard: Monthly Renewals Health; Business question: which accounts are at risk before renewal; Audience: customer success managers, VP Sales; Filters: region, plan tier, renewal month; Refresh: daily.