“Are your customers, partners and employees complaining about how slow your applications are running?" If you are not worried about the answer to this question, stop reading now.
Has your website stopped accepting appointments? Have your customers lost access to their bills? Is your intranet sluggish? Are there no more online sales? Are users complaining?
Brand damage and customer turnover are not the only consequences of low-quality customer experience. Costs also increase. Outage incidents, extended loading and display times, and overall service degradation have direct and indirect impacts on operations. Beyond user dissatisfaction, manually processing a transaction that could not be processed online disrupts workflows and undermines the efficiency and effectiveness of your colleagues.
You probably already have a large assortment of tools that monitor the performance of applications, networks, and cloud platform infrastructure. You may even think you have too many! The problem is, however, that these tools are almost exclusively focused on internal or backend issues. And, given the complexity of modern digital environments, it is just about impossible to infer what an end user or customer is experiencing from the data captured by backend-oriented performance monitoring technology.
So, how do you acquire timely and accurate insight into the digital experience of your end users and customers? There is, in fact, a solution that is simple, efficient and quick to implement: synthetic monitoring. You start by straightforwardly building user journeys or scripts and then hand them over to a population of software robots located at strategic points around the world. At regular intervals, these robots run the scripts against your user and customer-facing applications and record the results, simulating, in a sense, the ways in which users and customers interact with your services. In the first instance, the data they obtain answers the question 'Is the application or service available or not to my users or customers?' If not, you will be alerted to the need for a quick response and remedy. Beyond that, the robots will ascertain the time it takes to execute a user journey and, hence, provide you with a first-level evaluation of application performance. It should also be noted that, while a comprehensive understanding of service behaviour requires input from those technologies that burrow deep into your digital environment, synthetic monitoring is often the only way to determine the impact of those infrastructural components that do not belong to you (e.g., DNS, CDN, third-party advertising services.)
Of course, to get a fuller picture of the digital experience, you need something more than the robots can deliver. Each of your users and customers has a unique usage context. You may access your services from an old PC on a shipping dock at the other end of the world. Another may be sitting in London or Paris but has to cope with a degraded connection. Yet another may be accessing your services over high-speed lines with the latest equipment but needs to negotiate a complex array of security barriers. Furthermore, even if your DevOps team has rigorously tested the code, it was most likely tested in an ideal environment, giving you a limited grip on how that code behaves 'in the real world'. So how do you take the next step in digital experience monitoring? Here again, the solution is simple, implement Splunk's Real User Monitoring (RUM). Installation is easy and requires only the minimal intervention of introducing a tag into the application or service code. With the tag in place, then, you can capture specific measurements for each user session via the web browser.
Within its Observability Cloud Suite, Splunk's RUM solution has the ability to record navigations to reconstruct them as session replays (a video recording that reconstructs what your user saw; you relive their session). Here are some situations where they come in handy.
Reconstruct and visualize user sessions for faster diagnosis.
Life would be too easy if we could predict the future and know when the next incident will happen and what it will be! When a user reports an error or unexpected behaviour, your development team can use the replay facility to see exactly what the user did before the error or odd behaviour occurred. This helps them understand the sequence of events that led to the problem. It is like being right there with the user, seeing what he says, so that they can better solve the problem. This is particularly helpful when dealing with incidents that are rare or unique since these incidents can now be recorded and preserved for further analysis of their origins.
Replays can be used to validate that test scenarios are correctly implemented. Developers can review sessions to ensure that all test steps are executed correctly.
Before deploying an update or new feature, developers can use session replays to ensure that everything works properly and that the new feature doesn't introduce any problems.
Testing is a wonderful thing but your users and customers are very creative! No matter how far-ranging the set of testing scenarios might be, consumers of your services will interact with them in unforeseen, novel ways. RUM will allow your developers to capture these creative outbursts and share them with quality teams to review and refine testing scenarios.
Session replays allow you to see how users interact with an application or site. Teams can identify slowdowns, excessive loading times, or interactions that cause delays, which are essential for performance optimization.
Because a video is better than a long speech... or a report full of technical indicators and other Core Web Vitals, you can show business teams proof in images of the impact on the experience of that high-definition (but ultra-heavy) image, that advertising banner, or that tag they asked to add (but which only responds half the time and slows the loading of other elements).
One of the concerns when using replays is user data privacy. They can record sensitive information such as passwords, credit card numbers, or other personal data. As shown in the screenshot below, Splunk RUM Session Replay masks and/or excludes this data during playback. Texts and images can be replaced by the symbol * to mask the data.
Texts and images are replaced by the symbol * to mask the data
Using the Session Replay function with Splunk Real User Monitoring offers valuable benefits to understand and improve the user experience more quickly.
Sign up for a free trial, or visit the Splunk Real User Monitoring product page for more!
This article was written by Arthur Hamon, who is part of the Splunk Observability Strategist team for the EMEA region. He is based in Paris, France (and in Brittany whenever he can).
The Splunk platform removes the barriers between data and action, empowering observability, IT and security teams to ensure their organizations are secure, resilient and innovative.
Founded in 2003, Splunk is a global company — with over 7,500 employees, Splunkers have received over 1,020 patents to date and availability in 21 regions around the world — and offers an open, extensible data platform that supports shared data across any environment so that all teams in an organization can get end-to-end visibility, with context, for every interaction and business process. Build a strong data foundation with Splunk.