Derived health
Availability and performance are separate signals.
For every monitor type, WebGuard derives availability from raw check evidence—such as the HTTP response, transport outcome, connection result, observed DNS values, heartbeat, or server metrics. HTTP and Keyword monitors can additionally define a response-time threshold and consecutive-check confirmation count.
- A slow but successful check creates a separate degraded-performance alert; it does not become a public availability incident.
- The configured consecutive-check count confirms a latency change before WebGuard alerts.
- When response time returns below the threshold, WebGuard sends a recovery notification through the selected channels.
At a glance
- Checks for required response text
- Supports HTTP request configuration
- Useful for content, checkout, and login confidence
How it helps
Use keyword checks for pages where a plain HTTP status is too weak, such as customer portals, checkout pages, or rendered marketing content.
WebGuard can combine keyword validation with expected HTTP status ranges, request timeouts, methods, headers, body, and authentication.
When the keyword disappears, the incident flow and notification channels handle the failure like other monitor types.