Beyond sufficiency: Necessary conditions for employees’ intention to use personalized learning systems

• Previous insights based on sufficiency logic are expanded by necessity logic. • Performance expectancy, effort expectancy, and trustworthiness are necessary. • Subgroup analyses with multiple NCAs reveal differing necessity effect sizes. • The present study advances theory development in the field of technology adoption. • The bottleneck tables provide tangible guidance for practitioners. Personalized learning systems offer tailored educational resources to individual learners, promising increased efficiency and engagement in workplace learning. While prior research has identified drivers of employees’ usage intention to use such technology from a sufficiency perspective, less is known about necessary conditions. Building on a previously published study that applied structural equation modeling, this paper reanalyzes the data of 331 employees by means of necessary condition analyses (NCAs). Thereby, the conventional probabilistic sufficiency perspective ('X produces Y') is supplemented by a necessity perspective ('X enables Y'). Utilizing the emerging NCA method uncovers essential enablers, meaning the behavioral intention typically does not occur in their absence. Thus, the present study expands previous findings on technology adoption through a necessity lens. To gain a more nuanced view, NCAs are conducted across subgroups (based on UTAUT2 moderators) to identify potential differences. The findings reveal that the two most prominent factors in technology adoption research, namely performance expectancy ( d = 0.12) and effort expectancy ( d = 0.17), are necessary for employees’ usage intention. Moreover, a lack of trustworthiness hinders performance expectancy ( d = 0.13) and, subsequently, the intention to use. These novel insights contribute to theory development and inform practitioners who aim to implement personalized learning systems.