Session replay in Motadata ObserveOps records what a real visitor did in the browser and plays it back step by step, with grouped JavaScript errors, Core Web Vitals and Apdex scoring, and a link through to the backend trace behind any slow call.
Session replay is a recording of what a real visitor did in the browser during a visit, played back step by step so a team can watch the interaction instead of interpreting a description of it. Motadata ObserveOps captures the session directly from the browser, recording sessions, page views, user actions, JavaScript errors, long tasks, and resource timing, and replays that activity in sequence, so a vague bug report turns into an observable path of clicks, scrolls, and page transitions.
A recorded session never stands alone. It sits alongside its JavaScript errors, its Core Web Vitals score, and the user journey and segmentation data collected for the same visit, so a replay carries the context needed to judge whether the experience was actually degraded.
Session replay works inside the Real User Monitoring module, which records the frontend browser experience only. When a recorded session points to a slow or failing backend call, Motadata ObserveOps links it to the related request through RUM-APM Trace Correlation, while the code path, service, and database call behind that request remain owned by Application Performance Monitoring.
Browser instrumentation records sessions, page views, user actions, JavaScript errors, long tasks, and resource timing directly from the visitor's browser, so a replay reflects the complete session rather than a sampled portion of it.
Recorded sessions replay in sequence, so a team watches the exact path a visitor took and the point where the experience degraded, instead of reconstructing it from a support ticket.
Session capture runs across Vue.js, Angular, React, and Next.js frontends, so replay is available regardless of which of these frameworks an application is built on. React and Next.js support arrived in the 8.2.2 release.
Errors captured during a session are grouped, so a replay is tied to the specific errors that occurred inside it rather than read against a separate, disconnected error log.
Every replayed session carries its Core Web Vitals (LCP ≤2.5s, FCP ≤1.8s, CLS ≤0.1, INP ≤200ms) plus an Apdex score against a two-second threshold, so a replay comes with a standard for whether the experience was actually poor.
Replays sit alongside the user journey view and can be filtered by geography, device, and network, so a recording reads in context and a pattern worth replaying can be checked against how widespread it is.
A W3C distributed trace context links a recorded session to its backend APM trace, so a replay pointing to a slow or failing call can be followed into the request that produced it.
Filter combinations, search queries, time ranges, and layout can be saved in Session Explorer so a recurring set of sessions is one click away. RUM data also feeds Custom Reports (Chart, Grid, Top-N, Counter, Aggregation), and RUM policies sit inside Unified Incident Declaration alongside Log, Flow, Trap, and APM policies, so a recurring frontend failure does not need a separate workflow.
Browser instrumentation records the session, including page views, actions, JavaScript errors, long tasks, and resources, across Vue.js, Angular, React, and Next.js applications. Each session carries the trace context that RUM-APM Trace Correlation joins to its backend APM trace.
Recorded activity is measured against Core Web Vitals and Apdex, and grouped by error, so patterns across sessions become visible rather than isolated events. RUM policies built on this data feed into Unified Incident Declaration alongside Log, Flow, Trap, and APM policies.
Session Explorer presents recorded sessions for replay, with Saved Views preserving filter combinations, search queries, time ranges, and layout, and RUM-APM Trace Correlation adding backend context where needed.
Session replay is not limited to a single frontend stack. Capture and replay work the same way whether an application is built on Vue.js, Angular, React, or Next.js, so a mixed frontend estate gets one consistent session recording and replay tool rather than a different one per framework, and the same Core Web Vitals thresholds and Apdex scoring apply no matter which framework produced the session.
Watching the actual session, scored against Core Web Vitals and Apdex, removes the need to recreate a fault from an incomplete description or judge it by feel.
Grouped errors, the recorded steps, and journey context point directly to the interaction that broke.
Replays show how visitors actually move through the application, not how a team assumes they do.
RUM-APM Trace Correlation links a recorded session to its backend APM trace, so the frontend and backend read as one story instead of two separate investigations.
Segmentation by geography, device, and network shows whether a problem is widespread or limited to a specific audience, and Saved Views keep that recurring investigation one click away.
Session replay owns one layer of the stack, what happened in the browser. When a recorded session shows a slow or failing call, the code path, the service it hit, and the database call it triggered are answered by Application Performance Monitoring, not by session replay. RUM-APM Trace Correlation is the seam between the two: it links a recorded session to its backend APM trace using W3C distributed trace context headers.
The same discipline applies to service-level tracking. A RUM Performance SLO can be evaluated against multiple frontend metrics at once, but the SLO capability itself belongs to the Service Level Objectives module, which builds the SLO from RUM's frontend metrics rather than functioning as a RUM capability.
Session replay in Motadata ObserveOps turns a vague report into a recorded, scored, and traceable sequence of what actually happened in the browser, evidence a team can act on instead of a description it has to interpret.
Discover how Motadata ObserveOps can help you monitor your infrastructure in real-time and respond to issues instantly.