Control Combinations in New Product Development Projects à Serge A. Rijsdijk and Jan van den Ende This article investigates the effects of and interactions among the three prototypical control mechanisms of outcome control, process control, and clan control. Outcome control refers to the specification and evaluation of desired outputs that employees should deliver. Process control concerns mechanisms that specify which behaviors are ap- propriate, and clan control relies on socialization processes among organizational members. The authors argue that, whereas previous research has generally treated the control mechanisms in isolation, the different control mech- anisms are interdependent and act as complements or substitutes in their influence on process performance, product concept effectiveness, and financial performance. The authors collected data through an online survey on 148 new product development projects conducted by high-tech companies located in the Netherlands. The results show that the combination of different controls may lead to synergistic or conflicting effects for the different new product development project outcomes. Outcome controls and clan controls synergistically increase process performance. This implies that the benefit of increasing goal clarity and project team autonomy through outcome controls can be strengthened by supporting it with social cohesion. In contrast, process controls hamper the effectiveness of clan controls in their positive effect on process performance. Potentially, the combination of these relatively information- intensive controls leads to a redundancy in communication that decreases process efficiency. In contrast to expec- tations, outcome and clan control only independently affect product concept effectiveness. Finally, the results show that outcome control and process control negatively interact in their impact on financial performance. This suggests that the installation of highly strict control systems that combine high levels of both outcome and process controls may lead to lower financial performance. The authors conclude that neglecting the interdependencies between different controls may lead to an incomplete insight into how firms can most effectively manage their new product development processes. Introduction A fundamental issue facing innovation manag- ers is how to exercise adequate control over individual product innovation projects in or- der to ensure that project goals are met. Enforcing control increases the chance that projects remain on track and that new products have a short time-to- market (Tatikonda and Rosenthal, 2000). Simulta- neously, however, managers may not want to limit the problem-solving capabilities of project team members that are needed to deal with the uncertainty and un- expected events that characterizes these projects. Previous research on managerial control has mainly focused on how managers choose their con- trol mechanisms (see, e.g., Harmancioglu, 2009; Jaw- orski and MacInnis, 1989; Kirsch, 1996; Snell, 1992), on the effects of individual control mechanisms on specific outcomes (Atuahene-Gima and Li, 2006; Bon- ner, Ruekert, and Walker, 2002; Cardinal, 2001; Jaw- orski and MacInnis, 1989), how these effects may be moderated (Carbonell and Rodriguez-Escudero, 2010; Koller and Langmann, 2008) or how they pay off in the front end of the new product development (NPD) pro- cess (Poskela and Martinsuo, 2009). A major issue that has received little attention so far, however, is how var- ious combinations of control mechanisms affect product innovation project outcomes. This is surprising, as sev- eral scholars have signaled that, in practice, managers apply different controls in parallel that may interact (see, e.g., Cardinal, Sitkin, and Long, 2004; Kirsch, 1997; Poskela and Martinsuo, 2009). The current paper therefore investigates the interde- pendency of the three prototypical control mechanisms of outcome, process, and clan control as distinguished by organizational control theory (Ouchi, 1979; Turner and Makhija, 2006). Outcome control refers to the specification and evaluation of desired outputs. Process à The authors acknowledge the Netherlands Organization for Sci- entific Research (NWO) and the Erasmus Research Institute of Man- agement (ERIM) for financing parts of this research. The authors also want to thank Sjahrina Alladin and Robert de Jonge for their help with data collection. Address correspondence to: Serge A. Rijsdijk, Rotterdam School of Management, Erasmus University, P.O. Box 1738, 3000 DR Rotter- dam, The Netherlands. E-mail: srijsdijk@rsm.nl. Tel: þ 31 10 408 2541. Fax: þ 31 10 408 9014. J PROD INNOV MANAG 2011;28:868–880 r 2011 Product Development & Management Association