The knowledge needed for development of new products is becoming increasingly complex and distributed. There is a need for firms to develop their capabilities for knowledge integration. Although prior knowledge-based literature has pointed to the importance of knowledge integration for competitive advantage, our understanding of how the integration of different types of knowledge affects new product development (NPD) performance is still limited. We quantitatively examine the effect of the integration of different types of knowledge on performance in NPD. Knowledge integration refers to the process of combining specialized but complementary knowledge. In this article we seek to study the effects of knowledge integration on NPD performance. Previous studies point to the positive effects of knowledge integration on NPD performance, but they do not distinguish between different types of knowledge in examining these effects. This article therefore seeks to contribute to this emerging literature by explicitly studying the integration of different types of knowledge and the effects that such integration have on NPD performance. We draw upon a classification of knowledge suggested by Ullman (1997) in discussing what types of knowledge that is particularly pertinent in engineering practices. That is, what knowledge engineers draw upon in conducting design and development work. We address the types of domain-specific knowledge, procedural knowledge, and general knowledge integration. Three hypotheses suggesting that the capabilities for integration of each type of knowledge respectively affect NPD performance positively are tested. A fourth hypothesis suggests that there are complementarity effects between integration of the three types of knowledge upon NPD performance. Data was collected during 2009 from a sampling frame of 355 medium-sized Swedish manufacturing firms in four industries. We received 193 valid answers, i.e. a 54% response rate) Hypotheses were tested with standard OLS regression together with EFA and CFA analysis. The results provide support for the first three hypotheses, while the fourth one was rejected. This implies that capabilities to integrate domain-specific, procedural and general knowledge are all independently affecting NPD performance positively, but no combined effect above and beyond the individual variables. This indicates that one knowledge type can be integrated without an absolute need to integrate two types or all three types.