Endpoint Monitoring

Endpoint monitoring observes the quality of the user experience from desktops and mobile devices (also referred to as endpoints) to proactively uncover performance problems and availability issues. Depending on the context, endpoint monitoring may refer to the following:

  1. Security monitoring of end-user devices
  2. Monitoring of application programming interface (API) endpoints
  3. Digital experience monitoring (DEM) of applications used by employees and end-users

Endpoint monitoring in the security context involves analyzing end-user devices to identify misconfiguration or unauthorized network access. Powered by machine learning techniques such as anomaly detection, the services categorized under this definition of endpoint monitoring focus on defending against malicious activity.

API endpoint monitoring refers to monitoring an application programming interface to gain insight into the availability and performance of the calls made to an API. Since modern applications are made up of microservices accessing backend data via APIs, monitoring the API endpoints has become the most reliable method of abstracting the performance monitoring of application infrastructure. We have devoted a separate article to explaining the various techniques used in API monitoring.

This article focuses on the third category referenced above, which defines endpoint monitoring in the context of digital experience monitoring (DEM). In this context, an effective endpoint monitoring solution must cover all aspects of the end-user experience to assist in troubleshooting device, network, and application issues. It should provide the ability to proactively monitor the experiences of users across all applications from any device.

Let’s break it down. The following is a table of the core components of a complete endpoint monitoring solution:

Component What It Is Why It Is Important
Application Performance Insight Visibility into users’ underlying application performance experience: latency, errors, network traffic, and availability Pinpoints performance issues affecting the user experience across applications
Device Performance Monitoring Visibility into the health and performance of user devices. These are commonly referred to as endpoints (e.g., work laptops) Similar to application performance insight, this functionality shortens problem resolution time
Network Monitoring Oversees active endpoints on the network to address VPN issues, network outages, device connectivity, and low network bandwidth Provides an easy way to organize telemetry data and identify network bottlenecks for fast resolution of networking problems
Third-Party Service Level Agreements (SLA) Analysis Analyzes SLA agreements to hold vendors accountable to optimal performance Identifies problems in third-party platforms affecting the end-user experience
User Behavior Analytics Provides insight into employee work trends: the most used applications, top device preferences, etc. Visibility into application usage helps focus IT resources on addressing the issues with the most impact
Self-Remediation Capabilities Gives employees visibility into intermittent network issues along with recommendations for remediation (i.e., troubleshooting steps) Avoids wasting IT resources on problems that can be mitigated without additional support intervention


The Need for Endpoint Monitoring

According to a 2020 OwlLabs survey, 70% of full-time workers in the United States worked from home, and 77% of respondents would like the option to work remotely after the pandemic ends. 

The office workplace has gone from cubicles in a building to a type of workplace that is location-independent, where employees interact with business applications over systems that are outside the company’s purview.

What has fundamentally changed is that employees can go to work by logging onto a Monday morning meeting via a mobile phone while taking the subway to the office, completing their latest assignments on their employee laptops during their lunch break at a coffee shop, and ending their day by sending an email from a tablet while in a conference room at the office.

The simplicity that used to exist in accessing applications across a single enterprise network has morphed into a dynamic web of public and private networks involving a virtual private network (VPN). When an employee connects online through one of their devices, commonly referred to as endpoints, many factors can affect the quality of the user experience.

This doesn’t just apply to employees: Any end-user accessing your business applications may be using one of dozens of permutations of access devices and network access points. Monitoring the quality of the modern user experience requires a new approach known as endpoint monitoring.

Components of Endpoint Monitoring 

An endpoint monitoring solution should be equipped with the capability to collect telemetry data from a user’s device, facilitate analysis of the data, and provide useful metrics to take action and resolve issues. Let’s take a closer look at how Catchpoint addresses these requirements.

Overall User Experience

Catchpoint provides a platform that is capable of analyzing user experience, endpoint (or device) performance, network performance, and application performance. The platform offers a dashboard view that provides a quantitative score for each of these metrics with the ability to drill down and filter events.The experience score measures the overall user experience and is computed by taking the average of the endpoint (device), network, and application scores explained in the section below.

Figure 1: Catchpoint Endpoint Monitoring Dashboard View

Device Performance Monitoring

The endpoint score is determined by factoring in CPU usage, memory usage, and WiFi strength. For example, if the user is accessing the company via VPN from a local coffee shop, the WiFi strength might be lower than in the office. With Catchpoint endpoint monitoring, it is easy to monitor the strength of the wireless network and proactively alert end-users of the issue.

Device issues can be difficult for users alone to act on and may require help from an IT organization. For example, unusual memory consumption on a user’s laptop machine can indicate that closing unused applications could improve performance, but it may not be practical for a user to take action to remedy a poor network connection or application issues. 

Catchpoint compares data from a user’s machine to a historical baseline and a threshold to help detect issues early on. The ability to triage problems based on empirical data helps many users become more self-sufficient in resolving simple problems and reducing their dependence on IT organizations. 

Network Monitoring

The network score is computed by factoring in the round-trip time (RTT) and packet loss metrics generated from ping tests and traceroutes. Latency and packet loss are also considered part of the network score. 

Catchpoint network monitoring facilitates the organization of all network devices you own by allowing you to assign human-readable names to the underlying IP addresses. This includes firewalls, proxies, servers, gateways, and other elements of network infrastructure. This improves device detection by allowing you to quickly identify known and unknown devices on your network. For example, from within the Catchpoint platform, you can add a range of IP addresses that should belong to a particular application and label them accordingly.

Additionally, you can set up and configure alerts that are based on metrics and a threshold value or range. You can specify whether the alert should be triggered based on location data and whether it should be endpoint-specific or application-specific. For example, an alert can be configured to go off when a user experiences poor performance across three or more of their open applications. This can notify a support team with critical metadata about what happened, enabling you to diagnose problems quickly (company outage or isolated incident) and act on them in a timely fashion.


Application Performance Insight

The application score is calculated using Real-User Monitoring (RUM) data. Specifically, it is scored based on how quickly a view becomes “visually complete” or the point at which all content has been fully loaded and is visible to users in the browser. This is a critical metric to help identify applications that are not behaving as expected.

Applications on a user’s machine emit troves of metric data that can paint a picture of how well applications are performing. How well an application performs is correlated with the number of errors, latency, and availability it experiences.

Collecting this data and viewing it on a dashboard facilitates the differentiation of user-centric issues from org-wide level issues at the application level. Determining the course of action to troubleshoot is streamlined by giving support engineers the ability to detect issues and resolve them rapidly.

User Behavior Analytics

Collecting and analyzing application usage data across the enterprise is required for optimizing the user experience, but it also unlocks other benefits. Companies can use this information to prioritize their application performance monitoring efforts by focusing on the most popular applications. They can also identify the use of applications with overlapping functionality and save on licensing costs.

Catchpoint enables customers to analyze how often and how much applications are used by employees. For example, take a popular application like Slack. Although Slack offers collaboration features similar to Microsoft Teams, some companies may, surprisingly enough, use both.This duplication of functionality may happen because the company’s engineering team adopted Slack before the rest of the company selected Teams as the standard corporate communication platform. Catchpoint enables enterprises to pinpoint duplication, measure the volume of traffic on each of the competing applications, and help save money in licensing costs.

Catchpoint collects application usage data from many popular applications from providers such as Atlassian, Google, Amazon AWS, Microsoft, and Slack.


Endpoint monitoring observes the quality and performance of applications from the perspective of end-users and employees. Given the complexity of public networks and third-party systems that separate end-users from the application infrastructure, traditional monitoring must be augmented with new solutions that monitor end-user devices, also known as endpoints.