Response Time Threshold Alerts

Get alerted when response times consistently exceed your limits.

Response Time Threshold Alerts

Set a per-site millisecond threshold. When response times exceed it for 2+ consecutive checks, you get a slow_response alert. Use this to catch performance degradation before it becomes downtime.

Setup

Go to Edit Site → Advanced Monitoring → set Response Time Threshold (e.g., 2000ms). Save. The threshold applies from the next check onward and does not affect historical data.

How It Works

• Response time is measured on every uptime check across all monitoring regions
• 2 consecutive threshold breaches trigger an alert (single spikes are ignored)
• The alert is delivered through all channels configured on the site
• A recovery notification fires when response times return below the threshold for 2 consecutive checks

Choosing a good threshold

Look at your last 30 days of response time history in the Performance tab. Set the threshold at roughly your p95 — high enough to avoid noise from normal variance, low enough to catch real degradation. Marketing sites typically use 2000-3000ms; API endpoints 500-1000ms.

Related settings

• Response degradation alerts — rolling-average baseline instead of a fixed threshold. Better for sites with variable traffic patterns.
• Performance budgets — page-level Core Web Vitals limits, separate from uptime response time.

FAQ

Does the threshold affect uptime %? No. The site is still considered "up" as long as it returns a 2xx status code. The threshold only controls the slow_response alert channel.

Can I set different thresholds per region? Not yet — the threshold is global per site. Vote for it on the roadmap.