Our study shows empirically that the choice of resource allocation strategy affects innovation performance. A policy of allocating resources to a broader range of innovation projects increases sales of new products, especially if these are truly novel, i.e. new to the market. The effect of greater breadth appears to outweigh that of increased resource allocation per project. We find further indication that the performance effect of breadth increases with commercial uncertainty. It is also stronger for firms that allocate resources more selectively at later stages of the innovation process. Based on these results, we theorize that breadth increases performance as it spreads a firm's bets on unproven innovative endeavors, and more so when these endeavors are more uncertain. Limiting resource commitments through selectiveness contains breadth's disadvantages, a combination that provides flexibility in resource allocation.