The Importance of Monitoring the User Journey
By: Janne Koponen, Principal Engineering Services Architect
Application performance monitoring practices are evolving quickly. More data is gathered for analysis now than ever before, and as new monitoring trends appear, integrating, normalizing and analyzing the increasingly complex data in context becomes more important than ever.
Not so many years ago it used to be good enough that the app was available. This meant getting simple performance metrics, uptime trends and alerts when things went wrong. Considering the amount of metrics coming in from the millions of users accessing portals every day, this was no simple task, but with good practices and the right solutions, problems were fixed swiftly to ensure uptime and functionality. Traditionally, this approach required specialized teams and tools. For example, a dedicated team would look after the storage cluster and the backend app servers, another team supported customers when issues occurred, and a third team investigated usage metrics to optimize app design and performance.
To achieve good monitoring practices, a host of monitoring, reporting, alerting, service desk and other solutions are leveraged to make sure that the flow of discovering and fixing issues is seamless. Today, this approach is no longer enough to stay competitive. Application performance monitoring needs a new model.
From performance metrics to customer experience metrics
App monitoring metrics are also evolving. The current, most common performance metrics collected from apps – such as error rates, average latency and availability – are mostly consumed by a limited group of people: the developers, testers, support personnel and app designers. Theses metrics are still important, but to really step into the shoes of a customer, more complex indicators are needed.
Let’s look at one APM metric – URL performance. To know that a URL delivers a web page quickly and reliably was, and still is, essential. URL functionality can be secured by looking at front end latency, errors and return code, and then by drilling into the backend App Server using solutions such as CA Digital Experience Insights. But what else does a URL signify?
Domain URL is the business front page – it is how customers access your offerings. Together with other URLs, internal or external, it constructs the functional page that users first see. On that page, there is a set of links, followed by other links set as a path to lead the customers to various offerings. The job of the first page and the subsequent path is to lead users to a completed business transaction which, for example, could be a purchase or subscription.
This is one user journey. In most apps, there are several different journeys depending on what choices customers make while navigating the app. To make these user journeys easy, we still need to collect URL and page performance metrics, but also data from user experience point of view. Where do users drop out? How many of the visitors complete a transaction? Did they click on the online help to chat with support?
Most monitoring tools today can help you make sure that the front page delivers what it is supposed to deliver in terms of performance, reliability and even design. CA’s new SaaS based digital experience monitoring solution, CA Digital Experience Insights, does more – it can provide insight into how many of those customers who visit the front page actually follow the path. The journey may include issues, such as slow performance, buttons that don’t work, third party links that crash – resulting sometimes to calls for online or telephone support. It might also involve long wait times when the user is wondering which link to click next, or simply, what to do. Purchase or not purchase? All these things tell us what the user is experiencing, and with these insights, we can help our customers make the right choice.
Looking toward the future, it is clear that the emphasis will revolve around proactive prevention of customer experience issues, which means much more intelligent predictive data analysis. Already in testing, user journeys need to be heavily simulated using modern app synthetic monitoring tools, such as CA App Synthetic Monitor, and the same principle of modern monitoring must follow in production.
User experience will become more and more important as a differentiator, perhaps even more so than price. This means that more data from various data sources is needed to support app monitoring. Scalable solutions are needed to collect, host and report on the increasing amount of data. Therefore, it is also extremely important to filter out noise, or non-essential alerts and metrics. CA Digital Experience Insights can help analyze user experience indicators, application performance metrics and infrastructure insights while doing just that.
Integrating different data sources will also be a key. If the web or the app server is down, traditional application monitoring tools are still needed to produce alerts immediately so that technical issues can be fixed quickly. However, user experience issues might also be caused by an external effect. To really understand a user experience issue, the accumulated data might also need to be normalized. To find trends, analyzing the data in context – meaning comparing it with other types of data from the same period – is very important. To generate effective and meaningful trend analysis, all relevant data, internal and external, must reside under the same roof, either by integrations, or by using a solution such as CA Digital Experience Insights, which uses scalable open technologies such as Kafka, Elasticsearch and Apache Spark.
CA Digital Experience Insights, through CA App Experience Analytics, also offer an extensive REST interface, to make it easy to also export data if it becomes necessary to integrate with third party tools and solutions.
Customer journeys can be extremely complex. Monitoring all these journeys, and ensuring a positive customer experience is the new challenge that app performance monitoring solutions face today. Good news is that with the right solutions and diligent app performance monitoring practices, the future seems bright for those who embrace them.