Meta-analysis of publications on the impact of digital marketing strategies on small and medium enterprise performance

Abstract

This study synthesises findings from 56 research articles published between 2010 and 2024 to examine how digital marketing affects the performance of small and medi-um-sized enterprises (SMEs). Using a random-effects model estimated with restricted maximum likelihood (REML), the analysis shows a clear and moderately strong positive relationship (r = 0.3715, 95% CI [0.3443, 0.3884], p < 0.0001). The lack of vari-ability across studies is remarkable, with an I² value of only 0.09%. This suggests a strong consistency in the observed relationship, even though the studies used different contexts and research methods. Mixed-effects moderation analyses show that the impact of digital marketing varies depending on industry structure. Specifically, service-sector companies see greater performance improvements than small and medium-sized manufacturing businesses. In contrast, the geographic setting, the research design, and the type of performance metric used do not significantly change the observed relationship. To assess the possi-bility of publication bias, the study used Egger’s regression test and funnel plot analysis. Neither method showed any signs of systematic bias. The combined results suggest that digital marketing is a strong and broadly effective way to improve the performance of small and medium-sized enterprises (SMEs), espe-cially in service-based industries where digital interactions with customers are crucial. However, the lack of diversity requires further investigation, as it could indicate hid-den similarities in how the studies were designed or how they measured things. Future research would benefit from more detailed studies over time, as well as a closer look at how different sectors or organizations might affect the results of digital marketing.