Alerts
System-generated alert notifications for anomalies in your energy data.
The Alertes page lives under Optimisations > Alertes in the sidebar. It shows alert notifications generated by the system when anomalies are detected in your energy data.
How alerts work
Detection [step]
The system identifies an anomaly (e.g., a consumption spike, missing sensor data, an efficiency drop).
Notification [step]
An alert is created and appears on the Alertes page.
Review [step]
You review the alert details to understand the issue.
Action [step]
You investigate the root cause and take corrective action.
You don't need to configure rules manually — the platform handles anomaly detection automatically.
Types of alerts
| Alert type | What triggers it |
|---|---|
| Consumption spike | Energy usage significantly exceeds the expected baseline for the time of day and day of week |
| Data quality | Sensor data is missing, intermittent, or reporting implausible values |
| Efficiency drop | A site's energy efficiency score decreases significantly compared to recent history |
| Anomaly | Consumption patterns deviate from expected behavior in other ways |
Alert priority
Critical
Immediate attention required. Examples: complete data loss from a site, extreme consumption spike suggesting equipment failure.
Warning
Should be reviewed within 24 hours. Examples: a sustained increase in overnight consumption, a single sensor reporting intermittent data.
Info
For awareness, no immediate action needed. Examples: a minor shift in consumption pattern, a sensor reading slightly outside normal range.
Reviewing alerts
Each alert shows:
| Field | Description |
|---|---|
| Type | The kind of anomaly detected |
| Priority | Critical, Warning, or Info |
| Site & sensor(s) | Where the anomaly was detected |
| Description | What was detected |
| Time | When the anomaly was first observed |
Review alerts regularly to catch issues early. A consumption spike caught on day one is much cheaper than one discovered on the monthly bill.
Relationship to Recommendations
Alerts focus on real-time and near-real-time anomaly detection. The Recommendations page provides deeper AI analysis with suggested actions. A critical alert may also surface as a recommendation with more context about root causes and remediation steps.