CNCF-hosted Co-located Events North America 2024

Running a few experiments is easy, especially when using feature flags. However, scaling up experimentation and adopting an experimentation-driven approach across an entire organization is much more challenging. How can the frequency of experiments be increased while maintaining quality and trustworthy results? This talk will address this question by exploring the methods top companies use to run their experimentation programs at massive scales. It will start with an introduction to experimentation-driven product development and discuss the advantages of these techniques. Then, the discussion will cover how experimentation programs can go wrong and, finally, how top companies structure their programs to solve these issues and unlock experimentation at a large scale.