Experimentation in Organizations: An Integrative Review

Organizations have long relied on experiments to guide decision-making. Yet, a comprehensive synthesis of this rich and timely empirical literature remains lacking. In this integrative review, we identify two primary streams of research (problem-solving-based experimentation and causal inference-based experimentation), which we organize using the classic variation–selection–retention framework. The problem-solving stream emphasizes iterative experimentation, learning from failure, and navigating organizational challenges, while the causal inference stream focuses on sharp identification, structured experimental designs, and bounded experiments, each rooted in distinct disciplinary traditions. Despite the differences, these perspectives offer complementary insights into how organizations experiment, learn, and adapt. By analyzing 177 empirical studies across several disciplines and integrating these parallel streams, we develop a unifying framework that highlights the key drivers, processes, and outcomes of organizational experimentation. We conclude by outlining promising avenues for future research, including deeper retention in shaping experimental effectiveness and organizational learning, overcoming cognitive biases, expanding the scope of experiments to strategy, organizational design, and people processes, and the possibility of the two streams to cross-feed: problem-solving to generate broad hypotheses that causal inference sharpens, and causal inference experiments to trigger reframing of the problem-solving experiments.