Date
Abstract
Machine learning capabilities are rapidly emerging as critical strategic assets, yet strategy research has not fully explored their implications for firms’ competitive advantage. This inductive, multicase study examines how characteristics of machine learning technologies shape the strategic flexibility and growth of ventures commercializing them. Through an inductive, multicase study tracking six early-stage machine learning ventures in an accelerator program, we reveal how their distinct phases of specialization and exploration are constrained by properties of the underlying technology. Our findings highlight a complex, perpetual interplay between training data, learning approaches, model architectures and applications over time. This interplay shapes ventures’ pathways, alternately expanding or limiting strategic options available to them. The study makes important theoretical contributions regarding the multifaceted, dynamic nature of competitive advantage derived from machine learning capabilities. It also carries practical implications for organizations seeking to build competitive advantage from predictive analytics and investors supporting them.