This study examines a two-year data collaboration between a major Korean logistics operator and a big-data firm as a single-case study. Resource complementarity serves as the starting point of the collaboration, where combining mutually lacking resources―operational data and analytical capabilities ―could generate multiple forms of business value. The study analyzes, within a single causal chain, how these values crystallize into sustainable data assets and how the settlement paths diverge under varying conditions. Pattern matching and causal-inference tests support analytic generalization. The findings confirm that, due to the non-rivalry of data, the same data resource realizes dual value when applied through distinct AI models. Furthermore, within this collaborative context, a multi-stage institutionalization of data assetization was observed, with the formal contract serving as a critical turning point. The study further distinguishes data friction into a data-analytic level and an institutional-environmental level, proposing an analytical framework in which differences across these levels correspond to divergent settlement paths for data assetization. From a practical perspective, the findings imply that even after business value is established, the final settlement of data assetization might be delayed if institutional-environmental friction remains unresolved.