Research-based spin-offs (RBSOs) are new ventures based on technology generated by public research institutions. The literature on academic entrepreneurship has devoted growing attention to the heterogeneity of RBSOs and their business models (Mustar, et al., 2006). Prior research on this topic has pursued a typology approach where RBSOs can only follow one business model at the same time. In this paper we challenge this assumption. Based on evolutionary theory one may expect that a venture experiment with new business models (variation), discharge unproductive ones (selection), and keep efficient models (retention). Nelson (1961) has dubbed the idea that a firm can pursue a several business models for the parallel-path approach. In order to examine the performance implication of parallel-path business models within the RBSO context, the following research question was asked: Does the technology transfer capability of RBSOs, as reflected in their innovativeness, depends on the type of business model and the number of business models they pursue?