When writing web performance tests, the tests you create from these tools come from lab data collected from pre-defined environments, devices, and network settings. Lab data allows you to reproduce performance results repeatably, making it useful for detecting and fixing performance issues early. However, lab data ignores a very important aspect: your users’ real experience of using your applications. Field data, collected from real users under various conditions, reflect how your application is being used in the real world and track the performance your users experience and even the errors they encounter. Real users also have different user behaviors, which cannot be all simulated realistically via test scripts or test cases. Real users use different devices, network conditions, caching mechanisms, and even system fonts that can impact how a site loads. Therefore, you must still complement your performance tools that are using lab data with field data or real-user monitoring (RUM) solutions to achieve a holistic approach to web performance testing. One way to do this is by complementing Grafana k6 browser with Grafana Faro.

In this talk, I want to introduce you to both frontend and backend performance testing and why a medley of these testing activities are needed to make sure that your website is performant. I'll also give a quick walkthrough as to how tools such as xk6-browser can help with running both protocol-level (how performance testing is normally run through concurrent interactions at the protocol layer) and browser-level tests (testing with real browsers to provide a more realistic performance test).By the end of this talk, you should be equipped with new knowledge regarding frontend and backend performance testing which you can apply to your work projects.