Objectively Gauging User Experience with Apdex and AppNeta

In today’s digital environments, slow is the new down. As users continue to grow increasingly accustomed to rapid response—and less willing to wait—slow application performance continues to be highly problematic for organizations. While downtime of major services can make headlines, it is more often slow performance that hurts businesses, ultimately stifling productivity and profits. In this post, we’ll explore how Apdex can offer a useful, objective way to measure application performance. In addition, we’ll outline how AppNeta by Broadcom delivers a range of capabilities that support effective measurement of Apdex scores and the user experience.

Introducing Apdex

User complaints often arise around their applications. This can include issues like pages not loading, excessive delays in response, or outright failure to access apps.

For network operations teams, it’s vital to understand the apps users are working with and how they’re performing. Teams must also be able to correlate that application information with what’s happening on the network side.

As teams look to effectively gauge application performance, Apdex has emerged as an important measure. Apdex is an industry standard formalized in 2005.  Over the years, Apdex has gained detractors, who tend to fault the simplicity of the scoring mechanism. However, because it enables teams to gain high level, fast comparisons of how applications are performing, Apdex remains useful for many teams, helping guide the prioritization of troubleshooting efforts.

In effect, Apdex seeks to assign a score to user satisfaction for load times. This score can provide a valuable way to measure the quality of the user experience delivered, and to objectively track how that measure changes over time.

How Apdex is calculated

The Apdex calculation is based on defining a threshold time in which a page load is satisfying and a time that users would tolerate. A standard threshold for satisfying page load times is four seconds or less, though this measure can vary by users, applications, and other factors. A tolerating page load time threshold could also vary, but in this example, we’ll say it is 12 seconds.

Next, each interaction is assigned a score. Each instance of a satisfying load time, in this case, four seconds or less, would be assigned a one. Any load times that would be in the tolerating category, that is, those that fall between four and 12 seconds, would be assigned a 0.5. Any load times longer than the tolerating time of 12 seconds would be considered a frustrated user experience and given a 0.

For any given time window, the Apdex score would be calculated by adding the score of one for each satisfied load time plus 0.5 for each tolerating load time, and dividing that number by the total number of interactions. (See calculation below).

The end result is a number between 0 and 1 (or a percentage between 0 and 100%), where zero indicates no users are satisfied and one (or 100%) means all users are satisfied.

Solution introduction: AppNeta

AppNeta delivers a range of capabilities that help teams gauge the user experience. The solution offers true end-to-end network observability across modern environments, eliminating visibility gaps introduced by the reliance on distributed, multi-vendor, multi-cloud networks.

With AppNeta, teams can spot not just downtime but slow performance. AppNeta enables administrators to actively monitor network delivery experience and leverage automated Apdex calculations to gain a high-level understanding of how web apps are performing. With the solution, teams can track the end-to-end delivery path of users, whether their connections rely on cloud networks, ISPs, SD-WAN technologies, or any mix of the above.

Here’s an introduction to some of the solution’s key capabilities:

• Multi-step interaction tracking. With AppNeta, teams can measure Apdex scores for specific hops on a network path. In addition, the solution’s experience module offers a milestones feature that enables teams to track Apdex scores for multi-step workflows that include multiple page loads.  For example, milestones can be used to model and track a specific user workflow, such as a user opening Microsoft 365, going to Excel, creating a spreadsheet, and saving it. This can provide invaluable intelligence. For example, administrators can see that an app service provider’s site is responsive, but that issues are arising after users log in and begin interacting with specific functionality of an app.
• Geographic insights. With AppNeta, teams can ensure they’re not missing the forest of widespread issues by being buried in the trees of isolated data points. The solution’s ability to provide geographic visibility is one way the solution does that. With AppNeta, an administrator can quickly see that users in a particular region are all experiencing slow performance, and then drill down to see that they’re all using the same ISP as their last-mile provider. In this way, they can quickly determine that the ISP is experiencing an issue.
• App visibility. AppNeta can provide visibility into the apps being used, and how they’re performing. With AppNeta, teams can quickly see that all users of a specific SaaS app are experiencing slow responses, and so quickly isolate the problem as one arising in the SaaS provider’s environment.
• Detailed, holistic view of metrics. In addition to Apdex, the solution can provide hop-by-hop visibility of specific activities in a given interaction, such as DNS resolution, HTTP redirects, SSL connections, and total round trip response time. (For more information on this capability, see our recent blog post, Enhanced Web Metrics: A Deeper Dive into Website Performance.) Operators can see load times for an entire page as well as for specific objects, offering granular insights. For example, this helps teams see if there’s an increase in the size of a particular object that is delaying total page load times. In addition, they can quickly see that a work-from-home user is experiencing an issue, and drill down to see that there’s a problem in the user’s home Wi-Fi network.
• Flexible, customizable dashboards. AppNeta offers complete customizability in terms of sorting and filtering. By default, the solution can display Apdex scores for a given milestone or transaction. Teams can customize the metrics displayed, filter based on specific tags, and more. With the solution, an administrator can view performance metrics for all cloud-based apps or apps running in a specific cloud provider’s environment. Teams can also track Apdex scores for milestones and specific network hops over different timeframes.
• Insights for fast identification and remediation. AppNeta helps teams quickly identify when issues arise and determine how to resolve issues that occur in internally managed networks. If issues arise in third-party networks, administrators can provide granular details to third-party service providers in terms of where the issue is arising. Network operations teams can then alert internal users about the issue and help provide workarounds.

Conclusion

Apdex represents a useful metric for many network operations and application teams. By providing a simple, objective way to gauge application response times, Apdex can help guide the prioritization of troubleshooting and optimization efforts. AppNeta makes it easy to track Apdex scores and leverage complete insights for gauging the user experience. With AppNeta, teams can establish baselines for user performance, track performance against those baselines, and, critically, determine how to improve these scores and the user experience. To learn more, be sure to view our Small Bytes webcast, How to monitor user experience with Apdex.