KPIs for Browser Monitoring | Germain Documentation

🎯 KPIs for Browser Monitoring

GermainUX provides preconfigured KPIs, measures, and pivots for analyzing browser performance, user experience, application behavior, and technical issues affecting real users.

🔎 Overview

These analytics help teams determine:

Question	Details
What users experienced.	What users experienced.
Which pages, interactions, or resources were slow.	Which pages, interactions, or resources were slow.
Which errors and browser problems occurred.	Which errors and browser problems occurred.
Who was affected.	Who was affected.
Whether the cause was related to the page, JavaScript, network, server, or a third-party dependency.	Whether the cause was related to the page, JavaScript, network, server, or a third-party dependency.
How frequently the issue occurred and its impact at scale.	How frequently the issue occurred and its impact at scale.

Preconfigured analytics are available after the applicable browser-monitoring component is deployed. They can also be customized for application-specific requirements.

⚙️ Monitoring components

Component	Role
GermainUX RUM JS	Continuously captures browser performance, user interactions, errors, network requests, memory, video, WebSocket, and other real-user data when the monitoring script can be deployed in the application.
GermainUX RUM Extension	Provides browser monitoring when RUM JS cannot be deployed directly or access to the complete browser-page context requires an extension. It also supports extension-specific capabilities such as browser-crash monitoring.
GermainUX JS Profiler	Provides deeper JavaScript function, browser, DOM, resource, error, and network evidence for code-level investigation.

The available KPIs depend on the deployed components, browser support, monitoring profile, application architecture, and enabled capabilities.

📊 KPI perspectives

Browser KPIs can provide one or both of these perspectives:

Perspective	Purpose	Examples
User experience	Describes what the user did or experienced.	Dead Click, Rage Click, Video Watched
Technology	Describes browser, network, JavaScript, or application behavior.	Page Memory, JavaScript Long Task, Outbound Fetch Request
Combined	Connects a user-visible problem with its technical context.	Error Click, Browser Crash, WebSocket Error

User interaction and click KPIs

KPI	Description
Dead Click	A click after which the page does not change, navigate, or scroll within the configured detection period.
Error Click	A click associated with a subsequent JavaScript console error.
Mouse Click	An individual mouse-click or touch interaction.
Rage Click	Multiple rapid and consecutive clicks on the same element, often indicating frustration or an unresponsive interface.
User Click	A user-interaction interval containing the synchronous and asynchronous activity occurring between consecutive clicks.

See Dead, Error, and Rage Click Monitoring.

Browser health KPIs

KPI	Description
Browser Report	Browser-reported issues such as crashes, policy violations, permission problems, and deprecated-feature usage.
Browser Crash	A browser, tab, GPU, extension, or forced-reload event detected by the RUM Extension.
JavaScript Long Task / Browser Freeze	A period during which browser main-thread activity exceeds the configured long-task threshold.
JavaScript Console Event	A captured JavaScript console error, warning, informational message, or debug event.
Unhandled Promise Rejection	A rejected Promise for which the application did not provide a rejection handler.
JavaScript Popup Dialog	An alert, confirmation, or prompt displayed to the user.
Page Not Found	An application-specific page-not-found event.

Core Web Vitals and page-performance KPIs

KPI	Description
Largest Contentful Paint	Measures when the largest image or text block visible in the viewport is rendered.
Element Performance Measure	Measures the rendering of an image or text element annotated by the application for Element Timing monitoring.
Custom Performance Measure	Measures an application-defined duration between two browser performance marks.
Server Timing	Captures server-side timing metrics exposed through an HTTP `Server-Timing` response header.

See:

JavaScript and page-memory KPIs

KPI	Description
JavaScript Heap Usage	Percentage of the available JavaScript heap currently in use.
JavaScript Heap Used Size	Amount of JavaScript heap memory currently used.
Page Memory	Memory consumed by the monitored page.

These measurements can help identify excessive memory usage and sustained growth that may indicate a potential memory leak.

Use the GermainUX JS Profiler when deeper JavaScript execution and code-level evidence is required.

See JavaScript and Page Memory Monitoring.

Network-quality and device KPIs

KPI	Description
Network Grade Changed	A change in the assessed quality of the user’s network connection.
Poor Network Location	A location sample captured when the client’s network grade becomes poor.
Battery Status Changed	A change in the device’s reported battery status.

See Battery Level Monitoring.

Availability depends on browser support and the permissions granted to the monitored page or extension.

Network-request KPIs

KPI	Description
Outbound Document Request	Measures full-page loading, including document and DOM-processing durations.
Outbound Fetch Request	Measures an outbound request initiated through the Fetch API.
Outbound CSS Request	Measures the loading duration of a stylesheet.
Outbound Image Request	Measures the loading duration of an image.
Outbound Script Request	Measures the loading duration of a JavaScript resource.
Outbound Resource Request	Measures the loading duration of another monitored page resource.

These KPIs help determine whether a poor page experience was caused by the document, an API request, image, script, stylesheet, server, or another dependency.

The JS Profiler can provide deeper JavaScript and resource-loading evidence for selected investigations.

See Network Request Monitoring.

Chat and CTI KPIs

KPI	Description
Chat Event	Captures configured chat activity, such as agent availability or messages sent and received.
Chat Transaction	Measures chat activity such as conversation duration or agent-response time.

Chat analytics can be correlated with the surrounding application experience and business workflow to determine whether friction came from the conversation, agent availability, chat technology, or another application condition.

See Chat and CTI Monitoring.

Video KPIs

KPI	Description
Video Event	Captures actions such as play, pause, seek, full-screen change, error, and completion.
Video Watched	Measures watched duration.

These KPIs support engagement, completion, abandonment, and playback-error analysis.

See Video Monitoring.

WebSocket KPIs

KPI	Description
WebSocket Connection Event	Captures an opened or closed WebSocket connection.
WebSocket Error	Captures a WebSocket error.
WebSocket Message Size	Measures the size of messages sent or received.
WebSocket Session	Measures the duration between the opening and closing of a connection.

See WebSocket Monitoring.

Content and message detection

KPI	Description
Phrase of Interest Appeared	Captures a configured sentence or text pattern whenever it appears during a monitored user session.

Whenever GermainUX needs to detect and analyze a specific sentence each time it appears in a session, create a Phrase of Interest for that sentence. The resulting event becomes available for KPI analysis, searching, segmentation, correlation, and Session Replay context.

See Custom Insights.

Website audit, SEO, and crawler KPIs

KPI	Description
Website SEO Audit	Captures website issues involving JavaScript, CSS, DOM size, memory, network performance, and supported third-party technologies.
Website URL Availability	Captures URL availability results produced by the GermainUX website crawler.

These KPIs require the corresponding website-audit or crawler capability.

Custom browser KPIs

GermainUX can collect and analyze application-specific browser data:

KPI type	Description
Browser Event	A custom discrete browser event, such as `Payment Failed`.
Browser Metric	A custom numerical browser measurement.
Browser Transaction	A custom activity with a start, end, and duration.

Custom browser data can be used to monitor application-specific errors, milestones, workflow steps, performance measurements, or business outcomes.

See Custom Insights.

Pivots

Pivots segment KPI results by available dimensions. Common pivots include:

Application, page, and component.
Request URL, resource type, and domain.
Browser, browser family, operating system, and device.
User, role, group, and department.
Session and correlation identifiers.
Error, status, severity, and Phrase of Interest.
Network grade and client location.
System, host, target, and environment.
Campaign, landing page, exit page, and referrer.
Video action, chat event, and WebSocket state.
Application-specific metadata collected through Metadata Extractors.

Pivots help determine whether an issue is widespread or concentrated within a particular page, resource, browser, location, user population, or application version.

Measures

Depending on the KPI and its data model, available measures can include:

Count and distinct count.
Minimum and maximum.
Average and total.
Current or latest value.
Duration and active duration.
Error or failure count.
Resource size and memory usage.
Throughput and request rate.
50th, 90th, 95th, 99th, and 99.9th percentiles.
Standard deviation.

Percentiles are particularly useful for page, request, and interaction durations because an average can hide severe degradation affecting a smaller population.

Example analyses

Business or technical question	KPI	Measure	Pivot
Which pages generate the most frustration?	Rage Click	Count	Page title
Which scripts are slowest?	Outbound Script Request	95th-percentile duration	Request URL
Which browsers encounter the most JavaScript errors?	JavaScript Console Event	Error count	Browser family
Which pages consume the most memory?	Page Memory	Maximum	Page path
Which locations experience poor network quality?	Poor Network Location	Count	Client region
Which videos have the lowest engagement?	Video Watched	Average duration	Video title
Which WebSocket connections fail most frequently?	WebSocket Error	Count	Application or page
Which custom business errors affect the most users?	Browser Event	Distinct users	Event name

Teams can drill from aggregate results into affected segments, individual events, and user sessions. Session Replay and the available JS Profiler evidence can then help explain what users experienced and why.

Customization

Preconfigured browser KPIs can be used without customization. Configure additional analytics only when the application requires them.

GermainUX supports custom:

Browser events, metrics, and transactions.
Performance marks and element measurements.
KPI constraints and definitions.
Measures, formulas, and pivots.
Phrases and messages of interest.
Static and statistical SLAs.
Correlation and tracing.
Watches and automated actions based on detected insights.

A Watch notifies GermainUX users about an existing insight or monitored condition. It does not create a new insight to be collected.

Deployment and configuration

For your browser environment
Review Browser Monitoring.
Review the Browser Monitoring Deployment Overview.
Deploy GermainUX RUM JS when the monitoring script can be added to the application.
Deploy the GermainUX RUM Extension when direct script deployment is unavailable or the complete browser-page context requires an extension.
Deploy the GermainUX JS Profiler when deeper JavaScript and browser diagnostics are required.
Review the available Browser Monitoring configuration.
Review the detailed Web Analytics KPIs, Pivots, and Measures.
Review the KPIs for User Monitoring and Replay.
Review the Browser Monitoring APIs.

ℹ️ Get Help

The Germain Team can help you set this up. Contact GermainUX Support.

Component: Engine, JS Profiler, Mobile, RPA Bot Recorder, RUM JS, RUM Extension

Feature Availability: 2017.1 or later