Purpose - In recent years, it has become extremely common for numerous brands to allocate the majority of their online marketing budgets to social media promotion projects led by internet influencers. However, to date, there remains a significant gap in relevant academic research: the complete conversion logic through which consumers transform from initial exposure to these promotional contents into the tangible revenue that brands can ultimately obtain has never been clearly sorted out. Drawing on the Elaboration Likelihood Model and Source Credibility Theory, this study specifically targets Generation Z consumers and sorts out a continuous mediating transmission path: starting from users' first exposure to promotional content while browsing social media, consumers' active engagement with the content is first triggered, followed by the gradual cultivation of purchase intention, which ultimately translates into the tangible Marketing ROI that brands can track. In addition to sorting out this conversion path, this study also aims to verify one question: can an influencer's own credibility act as a boundary condition moderating the entire path, affecting the strength of the correlations between every link in the chain.

Design/methodology/approach - This study adopts a cross-sectional questionnaire survey design, and a total of 251 valid questionnaires were collected from Generation Z respondents. All respondents were aged between 18 and 27 years, and were recruited through the [purposive/quota] sampling method within the restricted [context/geography] scope. All relevant variables used in this study were measured using mature multi-indicator reflective scales that have been validated in previous academic research, and all items in the questionnaire collected respondents' answers using a 7-point Likert scale. Both the measurement model and structural model of the study were tested using the partial least squares structural equation model (PLS-SEM) in SmartPLS 4 software. At the same time, 5000 bootstrap subsamples were extracted to verify whether the indirect effects and conditional effects in the preset path are statistically significant.

Research Findings: PLS-SEM analysis of 251 valid responses supported all five hypothesised paths. Social media exposure positively predicted consumer engagement (β = .363, p < .001; H1), and influencer credibility significantly strengthened this relationship (β = .162, p < .001; H2). Consumer engagement, in turn, was the strongest driver in the model, positively predicting purchase intention (β = .536, p < .001; H3), and purchase intention positively predicted perceived marketing ROI (β = .531, p < .001; H4). Influencer credibility also significantly strengthened the engagement–intention relationship (β = .170, p < .001; H5). The model explained 36.2% of the variance in consumer engagement, 38.5% in purchase intention, and 28.2% in perceived marketing ROI, confirming that social media exposure translates into perceived marketing ROI through the serial pathway of consumer engagement and purchase intention, with influencer credibility acting as a boundary condition that strengthens both the exposure–engagement and engagement–intention links.

Originality/value- This study brings three core breakthroughs to the existing academic achievements in the field of influencer marketing. First, the vast majority of previous studies in this field only stayed at observing short-term proximal attitude changes in consumers after they viewed promotional content, such as shallow-level reactions like whether they developed a favorable impression of the product. This paper sets the brand-perceptible indicator Marketing ROI as the final outcome of the entire causal chain centered on consumers' behavioral changes, breaking the limitation of previous studies that only tracked consumers' intermediate reactions, and truly extending the end point of the research to the actual commercial revenue that brands can obtain from their marketing investments. Second, this study sets influencer credibility as a moderating variable that affects the strength of the core links of the path, rather than a variable directly used to predict the final outcome. It not only confirms that an influencer's credibility can affect the final outcome of the marketing campaign, but further clarifies under what conditions promotional exposure on social media can be converted into measurable improvements in consumer engagement, and then from improved engagement into improved purchase intention, practically answering the question of when conversion will occur, rather than only staying at the level of confirming that conversion does exist. Third, the research object of this study is limited exclusively to Generation Z. This group has been proficient in using various social platforms since childhood, so the research conclusions summarized from older age groups cannot be directly applied to the unique consumption behaviors of this group. By narrowing the research scope specifically to this group, this study eliminates the applicability gap of previous old conclusions when applied to the new group..