We’re Doing It Live: A Multi-Method Empirical Study on Continuous Experimentation

Journal article, 2018

Context: Continuous experimentation guides development activities based on data collected on a subset of online users on a new experimental version of the software. It includes practices such as canary releases, gradual rollouts, dark launches, or A/B testing. Objective: Unfortunately, our knowledge of continuous experimentation is currently primarily based on well-known and outspoken industrial leaders. To assess the actual state of practice in continuous experimentation, we conducted a mixed-method empirical study. Method: In our empirical study consisting of four steps, we interviewed 31 developers or release engineers, and performed a survey that attracted 187 complete responses. We analyzed the resulting data using statistical analysis and open coding. Results: Our results lead to several conclusions: (1) from a software architecture perspective, continuous experimentation is especially enabled by architectures that foster independently deployable services, such as microservices-based architectures; (2) from a developer perspective, experiments require extensive monitoring and analytics to discover runtime problems, consequently leading to developer on call policies and influencing the role and skill sets required by developers; and (3) from a process perspective, many organizations conduct experiments based on intuition rather than clear guidelines and robust statistics. Conclusion: Our findings show that more principled and structured approaches for release decision making are needed, striving for highly automated, systematic, and data- and hypothesis-driven deployment and experimentation.