Dashboard Alert and Annotation Rules
Make dashboards explain themselves.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Define rules that make dashboards explain themselves. Dashboard purpose and audience: [DASHBOARD_PURPOSE_AND_AUDIENCE]. Key metrics and their normal ranges: [METRICS_AND_RANGES]. Data refresh and delivery channels: [REFRESH_AND_CHANNELS]. Past problems (ignored alerts, unexplained spikes): [PAST_PROBLEMS]. Create a decision tree for each metric: 1. Is a change worth an alert? Define the thresholds: absolute limit, percentage change versus last period, deviation from a rolling baseline, and a minimum volume to avoid noise. 2. Severity: info, warning, critical, with different channels and response times. 3. Who owns the response and what is the first action? 4. Is the cause already known (a release, campaign, holiday)? If so, add an annotation to the chart rather than a new alert. 5. Is the data suspect (late load, missing source)? Mark data quality status separately from business alerts. 6. Suppression and grouping: avoid duplicates, quiet hours. Deliver an alert rule table (metric, condition, severity, channel, owner, runbook link), an annotation policy (who may add notes, required fields: date, what, why, author), examples of good and poor annotations, a weekly review of alerts (false alarms, missed events), and rules to retire alerts nobody acts on.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.