Structural equivalence, tie heterogeneity, and operational performance of tourism firms: a network-analytic and machine learning approach

Purpose This study investigates how structural properties of tourism firms, specifically structural equivalence and tie heterogeneity, predict the operational performance of tourism firms.Design/methodology/approach Data were collected from 311 tourism firms in Egypt using structured questionnaires to identify relationships with business partners and institutional stakeholders. Five supervised Random Forest (RF) models were then employed to classify firms’ operational performance based on their structural properties.Findings All five models discriminate above chance, providing evidence that structural equivalence and tie heterogeneity carry meaningful, although moderate, predictive information for operational performance. Business structural equivalence shows the strongest directional relevance for quality and responsiveness, while institutional tie heterogeneity shows the most consistent directional signal for efficiency, responsiveness, and innovation.Research limitations/implications The findings’ generalisability may be limited due to the unique regulatory and institutional environment of tourism firms in Egypt.Practical implications Tourism firms prioritising efficiency, responsiveness, and innovation should strategically cultivate institutional tie heterogeneity by engaging with diverse institutional stakeholders such as tourism authorities, regulatory bodies, and professional associations. Firms focused on quality and responsiveness should attend to their structural positioning within business networks relative to high-performing peers, as business structural equivalence carries more operational relevance than partner diversity alone.