Usability testing observes people trying to complete tasks with an interface. It helps identify where the task or presentation causes confusion. The goal is to learn about the interface, not to test whether the participant is good at using computers.
Prepare a realistic task
Instead of “click the search button,” ask “find an example that shows how grid columns share space.” That leaves the participant free to choose a route. Recruit people relevant to the intended audience and include different access needs.
Observe without solving it for them
Explain the session and obtain appropriate consent. Ask the participant to describe what they are thinking if that fits the method. Avoid pointing out the next control when they hesitate. Record where the task stops, what the person expects and what they actually do.
Turn observations into a testable change
A participant searching for “center text” but finding nothing suggests a vocabulary or ranking problem. It does not establish that every reader needs a new navigation layout. Make a focused change and check it again.
Exercise: observe one person finding and copying a reference example. Write three separate notes: the observed event, your possible explanation, and the change you would test.
Small qualitative sessions can reveal issues, but do not establish population-wide success rates. A/B testing answers a different kind of question. Read UX to place the session within the full task.
Sources
Reference 1 provide the technical details and current compatibility information.