Abstract
Purpose – We surveyed managers and other professionals in 142 US hospitality firms to investigate the links
among relative firm size, market strategies (e.g. broad cost leadership and global growth) and nonmarket
strategies. It helps explain how businesssize influencesthese dynamics within the hospitality industry. Previous
research has primarily focused on generic marketstrategies without adequately considering the nuanced impact
of firm size on performance outcomes.
Design/methodology/approach – The data are analyzed using PLS-SEM and machine learning (ML) methods,
specifically the glmnet algorithm for Lasso regression, to explore the relationships between firm size, market
strategies and performance. ML’s ability to manage high-dimensional data and nonlinear relationships provides a
nuanced analysisthatsurpassestraditionalstatistical methods, enhancing the accuracy and depth of our findings.
Findings – The results depict a negative link between broad cost leadership and firm performance and a positive
link between a global growth emphasis and firm performance. Firm size did not influence either marketstrategy
or performance, but nonmarket orientation fully mediated the relationship between firm size and performance.
Research limitations/implications – We assessed nonfinancial and financial performance with self-typing
scales; objective measures can help evaluate strategy-performance linkages through a different lens and
potentially reduce the influence of common method variance. Also, our assessment was limited to a crosssection of US hospitality firms. Additional work in the hospitality industry isrequired to identify and corroborate
nonmarket strategies at the firm, strategic group and industry levels.
Originality/value – The current study employs both PLS-SEM and machine learning. It reinforces
configuration theory by combining market and nonmarket methods as business performance indicators in
hospitality firms.