Usability testing is a study in which real users perform tasks in a product (or prototype) while a researcher observes where they get stuck and why. The fastest route from team opinion to facts.
The mechanics are simple: the participant receives a realistic task ("find and order…"), thinks aloud, and the researcher observes without helping. Five such sessions typically expose most serious problems – one of UX research's best-documented findings.
Moderated tests (a researcher guides the session) yield deeper "whys"; unmoderated remote tests give scale at lower cost. You can test anything: a paper sketch, a prototype, a live product, a competitor's product.
The most common mistake: testing after launch, when changes are most expensive. Test prototypes – a fix in Figma costs hours; in production, weeks.