Why companies cannot use carbon paper to develop products -A multiple case study of experimental product development processes

Abstract: To avoid developing products that customers do not want, many companies use iterative experiments in their product development processes. Today's existing frameworks regarding iterative experiments provide generic tools for how to conduct experiments within product development. However, limited research has been conducted on how individual company characteristics, i.e. contexts, affect the development process. This paper uses a qualitative multiple case study of eleven Swedish startups towards the aim to create further knowledge on how company characteristics, in terms of contextual factors, affect the way they conduct experiments in their product development process. The results show that companies with a higher level of maturity and previous innovation performance, experience, have a more structured experimental process and at the same time use an experiment type that conducts multiple experiments simultaneously whilst companies with a lower level of maturity have a more unstructured process and conduct sequential experiments. Smaller companies focus more on cost and criticality in their sequencing whilst larger companies have the capability to overlook the more critical tests in favor of testing features of a higher level of uncertainty. Startups with a physical product are more likely to conduct theoretical experiments using smaller sample sizes on blind test groups while companies with non-physical products create prototypes to test features on larger sample sizes that are aware of the experiments. Finally, companies with a high level of R&D intensity are more likely to test many alternatives within each experiment, set based, whilst companies with less intensive R&D investments focus on one subexperiment and alternative at a time, thus point based.