Synthetic tests help teams catch customer-facing issues before users report them. But even then, responding to failures can still be time-consuming. When a Synthetic Browser or API test monitor fires, an engineer must first determine whether the failure is a real regression or a test configuration issue. Making that determination requires manual review, and once an issue is confirmed, the team still needs to correlate logs, APM traces, infrastructure metrics, and deployment signals across multiple sources to figure out where the problem started.
AI-assisted triage is now available in Synthetic Monitoring through Datadog’s AI failure summaries and Bits Investigation. From there, you can determine whether a failure is worth investigating, review the evidence surrounding it, and launch or automate investigations that correlate telemetry data across Datadog. Instead of spending time gathering context, you can focus on finding the likely cause and beginning remediation.
In this post, you’ll learn how to:
Determine whether a failure is worth investigatingGet to a root-cause hypothesis fasterConfigure automatic or on-demand investigations by monitor criticality
Determine whether a failure is worth investigating
