How is Usability Testing Conducted?
Test subjects are generally recruited based on a set of screening questions and then invited to a research facility which is set up with the product to be tested. Alternatively, some videoconferencing capability (e.g. Skype, FaceTime, Zoom) may be set up to enable an interviewer to see and be seen by a respondent.
- In most cases monetary compensation, known as an incentive, is offered for cooperating in the research, but in some instances participants are simply given the actual product to keep.
- A typical session lasts from an hour to 90 minutes.
With the help of the client, the researcher/interviewer will establish a set of tasks and questions to administer.
- During the course of the project, wherever possible it is a good practice to rotate the tasks to avoid any bias that one might have on another.
- The interviewer should not provide any assistance (other than to clarify what is to be done) and should present the task in exactly the same way to each participant.
- If the task cannot be completed, reasons for failure will be identified and aid the client in making necessary adjustments to the product or website.
- By closely observing each user, the researcher may also make note of any body language or verbal expressions (indicating frustration, confusion, dissatisfaction or signs of pleasure or satisfaction) – all of which may be incorporated into a final report and recommendations.
Even with a relatively small number of subjects, many results and comments will tend to converge quickly and make evident whether any areas are in need of improvement.
It is a good idea to periodically perform usability testing to ensure that new users can readily accomplish tasks with your product or website as well as to determine if any modifications have made a difference.