Abstract

An important managerial concern when innovation is carried out by outside solvers is the extent to which those solvers choose to participate, and if so, the effort level they exert to solve a firm’s problem. At the same time, neither the solvers nor the firm can be certain about the number of participants when they make their respective decisions. Participation uncertainty implies that a contest may attract insufficient submissions, or even none, in which case the contest is canceled at a cost. We derive the optimal contest design that maximizes the best submission net the contest cancellation cost. Further, we analyze the effect of imposing a performance threshold to regulate solver participation and submission quality. Finally, we show that growing solver population does not always improve the best submitted solution, even in the absence of any costs associated with such a population expansion.