We examine how a firm designs capital allocation and managerial compensation schemes to motivate a privately informed manager to (i) engage in innovative activity to search for, and (ii) guide the firm to invest in, a new investment project. We show that relative to the first-best, the firm allocates too little capital and provides too few incentives for the manager to expend innovative effort; the manager may violate the NPV rule by investing the allocated capital in a project with negative productivity. We provide several novel predictions that help identify firms that are likely to innovate and managers who are likely to follow the NPV rule. We also show that uncertainty and incentive pay can be positively related.