It is completely human for UX practitioners to subconsciously allow their own values, attitudes or even mood to creep into their work when conducting research, affecting their observations during usability testing sessions and influencing which data they place greater significance on. To reduce the impact of this effect, include more than one observer in the testing session and share/calibrate findings in the form of a mini-debrief once the participant has left. This increases the likelihood of capturing more data, and mitigates the effects of the observers’ individual preconceptions. Training all observers prior to testing sessions and ensuring a common ground of specific behaviours to spot (e.g. by using an observation checklist) could also ensure that data collection is as accurate as possible.