N-best evaluation for academic hiring and promotion

How should we evaluate candidates for academic positions? Current evaluations privilege quantity of publications and journal prestige, creating perverse incentives. To avoid this issue, I propose an N-best policy: Hiring and promotion committees should solicit a small number (N) of research products and read them carefully as the primary locus of evaluation, using other elements like letters or job talks to contextualize this work. This policy aligns evaluation more directly with the goal of selecting scientists who produce high quality work.