I85PREP
⌘ K
New Relic APM Practitioner study materials
New Relic APM Practitioner study resources

New Relic APM Practitioner: Digital Experience Monitoring

New Relic APM Practitioner · Work through your exam, one topic at a time.

All guide sections

Digital Experience Monitoring

What this domain tests

This domain covers the end-user side of the platform: browser and mobile agents, synthetic monitors and session replay. Questions ask you to pick the right instrument for a question, interpret Core Web Vitals, and separate front-end symptoms from back-end causes.

Core rules

  • The browser agent reports events, not only charts: PageView, PageViewTiming (carrying LCP, CLS, INP, FCP), AjaxRequest, JavaScriptError and BrowserInteraction. The Pro+SPA agent instruments soft navigations; a plain browser agent under-counts page views in a single-page app.
  • Core Web Vitals thresholds: LCP good ≤ 2.5s, poor > 4.0s; INP good ≤ 200ms, poor > 500ms; CLS good ≤ 0.1, poor > 0.25. INP replaced First Input Delay as the responsiveness metric in March 2024.
  • Synthetic monitor types: ping (availability only, no JavaScript or content assertion), simple browser (loads the page, cannot click or type), scripted browser (Selenium-style script), API test (step-based or scripted HTTP). Monitors run from public locations or private locations for internal endpoints, and secure credentials keep secrets out of scripts.
  • Mobile agents report crashes, handled exceptions, network requests and session traces.
  • Session replay shows what a user actually saw; it is sampled and complements metrics rather than replacing them.

Worked example

A site records 20,000 page views in a day: 1,400 have LCP above 4.0s (7%) and 400 have INP above 500ms (2%). LCP is the bigger problem and it is a loading metric, so the fixes are server response time, image weight, CDN placement and render-blocking resources. INP is an interaction metric, fixed by breaking up long JavaScript tasks. Compressing images will not move INP.

Traps

  • Using a ping monitor to assert page content or JavaScript behaviour.
  • Treating synthetic timing as real-user experience when it comes from a clean data-centre network.
  • Blaming APM when the browser waterfall points at a third-party script.
  • Forgetting that session-trace sampling makes trace counts unrepresentative of traffic.