Digital Experience Monitoring
Synthetics
Synthetic tests simulate traffic: API tests (HTTP, SSL, DNS, gRPC, WebSocket, TCP, UDP, ICMP), multistep API tests that chain requests with variables and assertions, and browser tests that run real Chrome sessions through steps and assertions. Tests run from managed global locations or from private locations you host yourself as a container. Types are frequently confused in questions - SSL certificate checks, for instance, are an API test subtype, not a separate product.
Real User Monitoring
RUM instruments browser and mobile applications, capturing sessions, views, actions, errors, resources, and long tasks. RUM data can be sampled per session, and Session Replay records (or samples) sessions with configurable privacy masking of text and inputs. RUM correlates with APM when trace_id propagation is enabled, letting you jump from a slow user view to the backend trace.
Retention and correlation rules
RUM sessions, views, actions, errors, and resources are retained for 30 days, while aggregated RUM metrics remain queryable far longer in dashboards. Session Replay has its own enablement toggle per application, its own sampling rate, and 30-day retention. Synthetic test results support alerting through Assertions and can be watched by CI/CD pipelines.
Worked example: a browser test asserts statusCode == 200 and responseTime < 2000 on checkout; at the same time RUM shows elevated error rates for the same view, and the injected trace_id links the failing view to the APM span - that chain (Synthetic → RUM → APM) is the answer pattern for "how do you find root cause for a user-facing failure".