Social media algorithmic governance and knowledge processes: building decision-making effectiveness and knowledge sharing

Purpose This study aims to investigate how perceived social media algorithmic transparency (SMAT) and fairness (SMAF) influence knowledge processes and organizational outcomes in professional digital environments. Drawing on knowledge management (KM), trust and algorithmic governance perspectives, the study conceptualizes social media algorithms as socio-technical knowledge intermediaries that shape how individuals access, interpret and apply knowledge. Specifically, the study examines the trust-enabled pathways through which algorithmic attributes affect tacit knowledge application (TKA), explicit knowledge application (EKA), decision-making effectiveness (DME) and knowledge-sharing intention (KSI). Design/methodology/approach Data were collected from 297 used participants in the United States through Prolific, focusing on individuals who actively use social media platforms for professional purposes such as networking, learning and information seeking. Structural equation modeling using SmartPLS 4 was used to test the proposed relationships among algorithmic attributes, trust in algorithms (TA), knowledge application and organizational outcomes. Serial mediation analysis was conducted to examine indirect pathways. Findings The findings reveal that both SMAT and SMAF significantly enhance TA, with fairness exerting a stronger effect. Trust subsequently promotes both tacit and EKA. TKA positively influences DME, while EKA strengthens KSI. Furthermore, transparency and fairness indirectly affect these outcomes through sequential pathways involving trust and knowledge application, highlighting trust as the central mechanism linking algorithmic governance to knowledge outcomes. Research limitations/implications The study uses a cross-sectional design and self-reported measures, limiting causal inference. Additionally, the use of a generally used sample may constrain generalizability to specific organizational contexts. Future research may examine additional algorithmic attributes, such as accountability and explainability and validate the model across sectors, cultures and regulatory environments. Practical implications The findings provide actionable insights for platform designers, managers and policymakers seeking to develop trustworthy algorithmic ecosystems. Enhancing transparency and fairness can strengthen user trust, improve professional decision-making and foster knowledge-sharing practices. Organizations can also integrate algorithmically curated insights into formal knowledge workflows to support more effective KM. Social implications As algorithmic systems increasingly mediate professional interactions and information access, transparent and fair algorithms can promote more inclusive knowledge ecosystems, facilitate informed decision-making and encourage collaborative knowledge exchange in digitally mediated environments. Originality/value This study extends KM theory by positioning algorithmically curated social media platforms as active knowledge intermediaries rather than passive information repositories. By integrating KM, trust and algorithmic governance perspectives, it demonstrates how transparency and fairness shape distinct trust-enabled pathways for tacit and EKA, thereby advancing understanding of knowledge processes in algorithmically governed environments.