Traffic alerts provide analytical insights about significant changes in your site's traffic patterns. These alerts display annotations on the traffic analysis graph, helping you monitor trends across marketing channels and identify unusual spikes or drops.

The system uses machine learning-based anomaly detection to identify statistically significant traffic changes. It compares current traffic against the past 28 days of data, flagging variations that fall outside 99% of the normal distribution. This requires at least 28 days of historical data to establish patterns.

To view traffic anomalies:

1. Open Analytics panel and select Traffic
2. Choose date range from the dropdown menu
3. Click alert icons to view details

When viewing an anomaly, you'll see:

• Date of occurrence
• Traffic volume during the event
• For spikes: Top viewed pages (up to 2) and their traffic percentage
• For drops: Pages with decreased traffic (up to 2) and their percentage
• Contributing traffic channels

This information helps identify factors affecting your traffic, allowing you to make informed adjustments to your marketing strategy. For example, if Facebook traffic drops, you might experiment with different social sharing approaches.

You can provide feedback on anomaly detection by:

1. Clicking thumbs-up or thumbs-down under "Was this helpful?"
2. Adding optional comments
3. Submitting your feedback

The system continuously improves as it collects more data, helping you better understand and respond to traffic patterns on your site.

Related Articles

Previous Articles