Software development as an experiment system (keynote)

Abstract

Most modern software development activities are focusing on domains of emergence where experts cannot know a priori what kind of software provides value to users and customers. This is fundamentally different to traditional software engineering for large systems where a priori analysis by experts is used to identify requirements. While the latter is gaining a niche software category, developing and establishing development practices for domains of emergence is becoming significantly important and urgent. A major challenge is to find the right scope for software development. There are many options on what to deliver. Software practices are needed that help in determining what customers want and creating the right capabilities for them. In this talk I introduce an approach for steering software development towards the right scope by continuously conducting experiments. This includes systematically observing users’ behavioral responses to stimuli such as features. Insights from experiments directly influence frequent iterative deliveries. Success cases from industry show that such an experimental approach helps companies to gain competitive advantage by reducing uncertainties and rapidly finding product roadmaps that work.