Platform engineering as organizational strategy: platform gravity, friction, and the economics of scaling internal developer platforms

Abstract This article presents Platform Gravity as a conceptual framework for explaining why some internal developer platforms become organizational defaults while others remain technically sound but organizationally marginal. The framework argues that platform outcomes depend on the interaction among three forces: adoption, friction, and the standardization value created by organizational complexity. The paper treats platform engineering not as a tooling initiative alone, but as a socio-technical capability for changing how software organizations scale coordination, governance, and operational work. The article makes five contributions. First, it situates internal developer platforms within adjacent literatures on software delivery performance, socio-technical congruence, Conway’s Law, modularity, software ecosystems, platform governance, developer experience, and AI-assisted software engineering. Second, it defines platform friction as a latent construct capturing the cumulative resistance teams encounter when evaluating, adopting, using, and extending a platform, and it distinguishes friction from related ideas such as developer experience, switching cost, coordination burden, and trust deficit. Third, it advances stylized models for complexity, adoption, friction, and return on investment, explicitly treating them as heuristic devices rather than empirically calibrated laws. Fourth, it provides diagnostic tables, proxy metrics, and a friction operationalization approach intended to support repeatable organizational learning and future empirical work. Fifth, it presents two anonymized organizational vignettes as mechanism-revealing illustrations, while clarifying their provenance, limitations, and non-generalizable status. The paper also examines platform engineering in the context of AI-assisted software engineering. It argues that to the extent that generative AI lowers the marginal cost of producing code-like artifacts in some contexts, the relative value of governed defaults, ownership clarity, verification pathways, and operational legibility increases. This claim is advanced as a bounded inference and research agenda, not as a settled empirical fact. Overall, the article offers a platform-specific vocabulary for analyzing how internal platforms become paths of least resistance, how platform friction can be diagnosed, and why platform investment should be evaluated across cost, engineering-flow, and risk-reduction outcomes rather than through a single metric.