An important topic in the recent literature on firms’ innovation is the question of whether, and to what extent, firms which innovate once have a higher probability of innovating again in subsequent periods. This phenomenon is called the ‘persistence of innovation’. Although the literature has established that innovation persistence is indeed important from an empirical point of view, relatively little attention has been paid to identifying the reasons why this is the case. This study proposes that the differences in innovation strategies across firms are an important driving force behind innovation persistence, and analyses this issue using a panel database constructed from R&D and Community Innovation Surveys in Norway. Empirical measures of various innovation strategies are identified by means of a factor analysis. A cluster analysis is used in addition to a dynamic random effects probit model to extend the methodology adopted by prior studies, for the purpose to not only examine innovation persistence, but also determine how this persistence is influenced by innovation strategies. The results support the idea that the differences in innovation strategies across firms are an important determinant of the firms’ probability to repeatedly innovate. The study also distinguishes the effects of strategy differences on the persistence of product and process innovation in all firms, and within high-tech versus low-tech firms.