While there is significant interest in improving university?industry interaction, literature on the university side has tended not to focus on the characteristics of the personnel involved and has largely ignored the issue that there are differences between types of faculty member in their degrees of interaction. This question is especially relevant at regional level, as those faculty members who do interact with industry may show a preference for firms that are larger and technologically superior to those in the region. Most analysts, however, have tended to focus on the national level, particularly on those countries at the forefront of technological innovation. In the absence of any formal theory, we propose a two-step method to formulate the hypothesis that only selected faculty members interact with selected firms. First, we identify the type of faculty member who interacts with firms. Second, we examine whether this type of faculty member interacts with every type of firm. A test sample is drawn from the Valencian Community of Spain, a region with low absorptive capacity, where firms may show undesirable properties for universityi nteraction. The results allow us to challenge the view that certain individual universities may show a higher propensity for interaction once we take into account differences between the individual characteristics of their faculty members. We also claim that in a region like the Valencian Community, faculty members who usually participate in contracts (male, holding an administrative position) do so mainly with larger firms, but not with firms from their own region, where they find lower technological standards. This partly explains the delocalisation of university?industry interaction.