Usability testing involves evaluating the ease of which a user or group of users can interact with a site, software application, etc. It is typically (although not exclusively) achieved by observing users as they attempt to complete a pre-determined set of tasks, with the aim of identifying any difficulties which may arise or any aspects that users particularly approve of; as well as generally noting any observations they might make. Data generated from testing can then be used to inform any subsequent design decisions.
In practice, usability testing is often an iterative process undertaken at various stages of a product's development cycle with prototypes of varying levels of fidelity. Employing a low-fidelity prototype of a website, for example, such as a paper sketch that approximates what it will look like, can be considerably more efficient than fully developing an actual website only to discover that it must be changed again.
