> ## Documentation Index
> Fetch the complete documentation index at: https://docs.fenra.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Alerts

> Get notified before AI costs spiral out of control

Alerts notify you when spending crosses a threshold. Set them once, and Fenra watches your costs 24/7.

## How Alerts Work

1. You define a threshold (e.g., "\$1,000 in 24 hours")
2. Fenra monitors your spending continuously
3. When the threshold is crossed, you get an email

No manual checking. No surprise bills at month-end.

## Creating an Alert

1. Go to **Alerts** in the navigation
2. Click **Create Alert**
3. Configure the alert:

| Setting         | Description                                                   |
| --------------- | ------------------------------------------------------------- |
| **Name**        | Something descriptive: "Daily Budget Alert"                   |
| **Threshold**   | Dollar amount that triggers the alert                         |
| **Time Period** | Window to monitor (hour, day, week, month)                    |
| **Filters**     | Optional: limit to specific models, environments, or features |
| **Recipients**  | Who gets notified                                             |

4. Click **Create**

<Tip>
  Start with a higher threshold and lower it as you understand your typical spending patterns.
</Tip>

## Time Periods

| Period            | Best For                            |
| ----------------- | ----------------------------------- |
| **Last Hour**     | Catching sudden spikes in real-time |
| **Last 24 Hours** | Daily budget monitoring             |
| **Last 7 Days**   | Weekly budget tracking              |
| **Last 30 Days**  | Monthly budget management           |

## Filtering Alerts

Make alerts more targeted with filters:

* **By Model**: Alert only when GPT-4 spending exceeds threshold
* **By Provider**: Separate alerts for OpenAI vs. Anthropic
* **By Environment**: Different thresholds for production vs. development
* **By Feature**: Monitor specific high-cost features

### Example: Production-Only Alert

To monitor only production spending:

1. Set your threshold (e.g., \$5,000)
2. Add a filter: Environment = "production"
3. Now staging and dev costs won't trigger this alert

## Alert Frequency

Control how often you're notified:

| Option        | Behavior                                 |
| ------------- | ---------------------------------------- |
| **Immediate** | Alert as soon as threshold is crossed    |
| **Throttled** | Max one alert per hour/day to avoid spam |

<Warning>
  Without throttling, a sustained overspend can flood your inbox. Enable throttling for high-frequency alerts.
</Warning>

## Managing Alerts

### Pause vs. Delete

* **Pause**: Temporarily stop an alert. Turn it back on later.
* **Delete**: Permanently remove it.

Pausing is useful during known high-spend periods (launches, migrations, etc.).

### Editing

Click any alert to update its threshold, time period, filters, or recipients. Changes take effect immediately.

### Alert History

See when alerts fired and why:

* Timestamp of each trigger
* Actual spend vs. threshold
* Who was notified

## Use Cases

<CardGroup cols={2}>
  <Card title="Budget Guardrails" icon="shield">
    Alert at 50%, 75%, and 90% of monthly budget for early warnings.
  </Card>

  <Card title="Anomaly Detection" icon="triangle-exclamation">
    Set a low hourly threshold to catch unexpected spikes immediately.
  </Card>

  <Card title="Feature Monitoring" icon="lightbulb">
    Watch specific high-cost features with targeted alerts.
  </Card>

  <Card title="Environment Separation" icon="server">
    Higher thresholds for production, lower for development.
  </Card>
</CardGroup>

## Best Practices

1. **Set multiple thresholds**: 50%, 75%, 90% of budget gives progressive warnings
2. **Use filters**: Generic alerts are noisy; targeted alerts are actionable
3. **Enable throttling**: Avoid alert fatigue
4. **Review monthly**: Adjust thresholds as your usage patterns change

## Related

* [Dashboard](/product-overview/dashboard): See spending trends
* [Cost Explorer](/product-overview/cost-explorer/breakdown): Investigate what triggered an alert
* [Reports](/product-overview/reports): Generate post-mortem reports