Turn vague data reliability expectations into measurable SLOs and error budgets.
#slo
#observability
#reliability
#error-budget
Prompt
Design service-level objectives for a critical data product. Focus on what consumers experience, not only whether orchestration jobs are green.
Data product and business purpose: [description]
Consumers and decisions enabled: [teams/use cases]
Critical datasets and dependencies: [assets]
Current update cadence and latency: [details]
Known incident history: [frequency and impact]
Business hours and critical windows: [e.g. market open, month close]
Existing telemetry: [freshness, volume, quality, lineage]
Deliver:
1. Consumer journeys and critical data indicators
2. Precise SLIs for freshness, completeness, correctness, availability, and change failure rate
3. SLO targets with measurement windows and exclusion rules
4. Error-budget calculations with worked examples
5. Multi-window burn-rate alerts that balance speed and noise
6. Dependency-aware alert routing and ownership
7. Policy for release freezes, reliability work, and budget exhaustion
8. Dashboard specification and weekly review format
9. Instrumentation gaps and an implementation plan
10. Examples of misleading metrics to avoid
Define every metric so two engineers would calculate the same value. Explain why each target matches business impact.
Customize
Fill Variables
Details
Model
GPT-4
Category
Data Engineering
Added On
Jul 12, 2026
Prompts are starting points. Review outputs before using them in production.