In the December edition of Research Policy there is an interesting article by Tavassoli & Karlsson (2015) about the persistence of innovation. The persistence of innovation can be described as the phenomenon where past and current innovation are a good indicator for future innovation because there is the belief that the competitive advantage of firms are rooted in the the capability to innovate over longer periods of time. In this context the persistence of innovation is sometimes referred to as the Innovation Marathon or Ultra Innovation which is derived from ultra running (running long distances past the marathon and often for several days in different sorts of terrain).
The research of Tavassoli & Karlsson (2015) shows that innovation is persistent. However product innovation is more persistent than process innovation, organization innovation and marketing innovation. Product innovation, process innovation and organization innovation have also a so called true pattern of persistence. This means there is a clear path-dependent process between the decision to innovate in one period and the probability to decide and to succeed to innovate in the following period. Product innovation shows the most clear path-dependency. Marketing innovation however shows no significant path-dependency as it is more variable because it can be stopped more easily if necessary without severe consequences.
The article shows that innovation is persistent (innovation leads to new innovations), whereby product innovation is most persistent and most path-dependent (true state). To my understanding from business practice it looks logical as there is a more direct relationship towards specific customer demand and customer expectations, that is driving the persistence of the innovation process.
Tavassoli, S., & Karlsson, C. (2015). Persistence of Various Types of Innovation Analyzed and Explained. Research Policy, 1887–1901.