Carbon monitoring systems generate continuous emissions data across facilities. The real challenge isn't collecting signals. It's helping operators act on them quickly and confidently. This feature redesign focuses on turning raw alerts into structured, actionable workflows. I focused specifically on severity modelling, alert lifecycle design, and resolution tracking.
Business and User Goals
Business Goals:
- Reduce compliance response delays
- Improve visibility across facilities
- Create structured audit-ready alert history
User Goals:
- Identify critical alerts instantly
- Understand context without switching views
- Resolve and document actions efficiently
Design Approach
The notification system was redesigned around three principles:
1. Structured Severity Model
- High – Critical emission spike
- Medium – System warning
- Low – Advisory update
- Escalation – Facility-level trigger
2. Action-Oriented Alert Cards
Each alert includes:
- Clear status badge
- Assigned owner visibility
- Primary action (Acknowledge / Resolve)
- Contextual metadata
3. Configurable Alert Settings
Operators can toggle:
- Emission sources
- Workforce activity
- Compliance tasks
- Sensor anomalies
- Facility grouping
Constraints
This feature was designed within an existing enterprise system handling live emissions data across multiple facilities.
Key constraints included:
- Real-time data streams that could not be modified at the source
- Legacy dashboard structure that limited layout flexibility
- Compliance workflows that required audit traceability
- The need to reduce alert fatigue without hiding critical events
The system had to improve clarity without introducing operational risk.
Flows (UI Screens)
The following screens illustrate the alert lifecycle, from detection to configuration and resolution.
Impact
- Reduced cognitive overload through structured prioritisation
- Improved clarity in high-risk compliance scenarios
- Structured resolution tracking for audit readiness
- Scalable system adaptable across facilities
Scope Decisions
To maintain focus and ship a stable workflow, the following were intentionally excluded:
- Predictive compliance risk scoring
- Automated escalation chains across departments
- Advanced analytics dashboards for executives
- Machine learning-based anomaly detection
The goal of this phase was structured alert handling, not full-scale compliance automation.
Key Trade-offs
Several trade-offs shaped the final solution:
- Chose a simplified four-level severity model instead of granular categorisation to reduce decision friction.
- Prioritised action-oriented alert cards over dense data tables to improve scanability.
- Limited configuration depth in the first version to avoid overwhelming operators.
- Focused on lifecycle tracking rather than predictive intelligence in this phase.
These decisions were made to improve operational clarity over feature breadth.
Future Enhancements
- Predictive alert grouping
- Executive summary dashboard
- Usability validation with compliance teams
Reflection
Designing for compliance-heavy systems requires balancing visibility with noise reduction. Clear prioritisation and structured workflows matter more than visual complexity.