Cloud Testing Blog

Key takeawaysID is fastest and most stable; CSS is the practical default when there is no ID.XPath is the slowest and the most brittle, especially when it depends on position.A ...

Key takeawaysEmulators catch layout and viewport problems quickly and cheaply.They do not reproduce real network or CPU conditions, so confirm on a device before shipping.Test the viewports in your analytics ...

Key takeawaysUI steps let you script a real journey rather than checking a single URL.Build the scenario in the order a user would follow, then run it to confirm each ...

Key takeawaysUse the Select class for real select elements, and normal clicks for custom widgets.Check the tag before choosing, since many dropdowns are divs pretending to be selects.Select by visible ...

Key takeawaysNon-GUI mode is the only sensible way to run a real load test.Pass the test plan and the results file on the command line and let it run headless.The ...

Key takeawaysPage speed is a ranking factor, so a slow site costs you traffic before anyone reads a word.Measure first with a real testing tool. Optimising without a baseline is ...

Key takeawaysgzip on NGINX is a few lines of config and usually the cheapest bandwidth win available.Compress text types only. Compressing already-compressed images wastes CPU for nothing.Set a sensible minimum ...

Key takeawaysThe XPath Extractor pulls values out of an XML or HTML response for use in later requests.Match on a stable element rather than position, which changes with the response.For ...

Key takeawaysWithout assertions a test passes on any response, including an error page returned as 200.Assert on content as well as status, since a wrong body is still a failure.Keep ...

Key takeawaysStart with a thread group, an HTTP request and a results file: everything else is refinement.Ramp gradually so you can see where behaviour changes rather than only that it ...