Conceptual goal models are used to express intentional aspects of the system under de-velopment. Among others, goal models facilitate stakeholder discussions and agreement about main system aspects during early requirements engineering phases. As experiences from participatory design indicates, the use of multimedia representations (especially videos) leads to better stakeholder involvement and, as a consequence, the produced con-ceptual (goal) models and specifications respectively are of higher quality. In this paper, we report on our empirical investigation which shows that the use of asso-ciations between goals and video parts documenting goal achievements and goal failures improve the performance of typical requirements engineering tasks. More precisely, they lead to more correct and complete results.