Presenting commercial goods through influencer-generated social media posts makes marketing content appear significantly less like an advertisement than official branded publications, yet this reduced commercial appearance does not translate into higher consumer attractiveness, greater product relevance, or stronger declared purchase activation. A laboratory experiment led by Róbert Štefko, Veronika Loumová, and Dávid Miško examined how consumers subjectively evaluate and visually process identical commercial items across different digital formats. By combining self-reported evaluations with objective eye-tracking metrics, the research team demonstrated that consumers distinguish perceived commercial intent from their overall evaluation of visual content. Although marketers frequently assume that influencer endorsements inherently outperform traditional brand communication by appearing more organic and authentic, the exploratory investigation showed that diminished advertising character does not automatically enhance consumer interest or purchase readiness when exposure conditions and presented products are held constant. The findings emphasize that visual marketing effectiveness cannot be understood as a uniform phenomenon.
The theoretical foundation of the investigation centers on how digital communication touchpoints shape consumer decision-making and cognitive processing during product discovery. Modern online shoppers rarely encounter products through a single isolated channel, instead navigating an interconnected landscape of official brand profiles, sponsored creator content, and detailed product webpages. Drawing upon the persuasion knowledge model, source credibility frameworks, and dual-process cognitive theories such as the Elaboration Likelihood Model, the authors highlighted that contextual delivery influences how persuasive messages are perceived. Official branded communication conveys explicit promotional authorship and firm-controlled commercial information, whereas influencer-generated posts blend promotional messages with personal experiences, authentic recommendations, and lifestyle storytelling. In contrast, web-based product pages provide information-dense environments that encourage systematic, deliberate evaluation during the information search stage. However, prior digital marketing literature has rarely compared visual attention and subjective responses to identical products across these three ubiquitous formats simultaneously within a single experimental framework.
To explore these mechanisms under controlled experimental conditions, the researchers recruited a sample of 30 university students enrolled in business, management, and economics programmes at a Slovak university. The participant cohort comprised 16 women (53.3 percent) and 14 men (46.7 percent) with an age range spanning 17 to 25 years, a mean age of 20.67 years, a standard deviation of 1.75 years, and a median age of 20 years. In terms of educational attainment, 23 participants (76.7 percent) had completed secondary education with a school-leaving examination, while 7 participants (23.3 percent) held a completed first-cycle university degree. Social media engagement was exceptionally high across the sample, with 29 participants reporting platform use several times per day and one individual reporting daily usage. Participation was voluntary and uncompensated, without course credit or financial payment, and suitability for eye-tracking was confirmed practically through individual calibration rather than formal visual acuity testing. No participant-level exclusions were required.
The experimental protocol utilized a repeated-measures design where each participant viewed 12 commercial visual stimuli displayed on a computer screen under standardized laboratory conditions. Four distinct products were each rendered into three naturalistic content formats: an official web-based product page, a branded social media post, and an influencer-generated social media post. Administered through iMotions 10.1 software, each visual stimulus appeared for an exact fixed duration of five seconds in a randomized sequence to prevent order biases, gradual task adaptation, or cognitive fatigue. Gaze behavior was recorded at a sampling rate of 60 Hertz using a screen-based Smart Eye AI-X eye tracker positioned approximately 60 centimeters from the participant. An individual nine-point calibration preceded testing, resulting in excellent calibration quality for 18 participants and good calibration for 12 participants. Eye movements were classified via default velocity threshold identification settings with a 20-millisecond window length and a 30-degree-per-second velocity threshold, without post-processing interpolation, noise reduction, or fixation filtering.
Immediately following each five-second stimulus exposure, participants completed a Slovak-language questionnaire block assessing their conscious reactions to the viewed material. Items Q1 through Q8 measured post attractiveness, perceived product relevance, declared purchase activation, and perceived advertising character across all conditions, while influencer stimuli included supplementary questions Q9 through Q11 evaluating interpersonal trust, perceived closeness, and willingness to follow the profile to construct a composite influencer affinity index. For physiological processing, raw eye-tracking outputs—total fixation time, fixation count, saccade count, and mean fixation duration—were aggregated across each respondent-stimulus presentation, yielding a full dataset of 360 observations containing valid fixation and saccade events. Standardized composite scores were derived to measure overall visual engagement, visual scanning activity, and a fixation-duration-based processing depth index, with internal consistency across aggregated participant-level measures demonstrating satisfactory reliability. The derived indices were treated as exploratory measures designed to capture visual dynamics rather than standardized clinical scales.
Repeated-measures analysis of variance revealed that the effect of digital content format was highly selective across conscious evaluative dimensions. The omnibus test identified a statistically significant difference among presentation formats exclusively for perceived advertising character, with an F-statistic of 5.62 across 2 and 58 degrees of freedom, an exact p-value of .006, and a partial eta-squared of 0.162, a conclusion corroborated by a non-parametric Friedman test yielding a chi-square of 7.51 (p = .023, Kendall’s W = 0.125). Conversely, presentation format exerted no statistically significant omnibus effects on post attractiveness (F(2, 58) = 2.61, p = .080, partial eta-squared = 0.083), declared purchase activation (F(2, 58) = 2.13, p = .130, partial eta-squared = 0.068), or perceived product relevance. Descriptively, branded posts registered the highest mean attractiveness, purchase activation, and commercial salience, whereas influencer posts were perceived as least commercial without securing an evaluative advantage. The lack of statistically significant differences across these evaluative outcomes highlights that lower commercial recognition does not necessarily boost purchase interest.
In examining objective visual behavior, the repeated-measures analysis of variance documented a statistically significant omnibus effect of content format solely on the processing depth index, which was derived directly from mean fixation duration, yielding an F-statistic of 3.93 across 2 and 58 degrees of freedom, a p-value of .025, and a partial eta-squared of 0.119. A complementary Friedman test confirmed this divergence with a chi-square of 8.60 (p = .014, Kendall’s W = 0.143). In contrast, global visual engagement and visual scanning activity demonstrated no statistically significant differences across the web, branded, and influencer formats. The authors emphasized that variations in mean fixation duration represent physiological fixation behavior within a fixed time window rather than unmediated evidence of superior cognitive comprehension, noting that longer individual fixations can easily stem from visual ambiguity, layout complexity, or heightened interpretive effort rather than heightened consumer appreciation. Format differences therefore manifested selectively in specific ocular durations rather than general scanning behavior.
Subsequent post hoc pairwise testing applying Holm corrections clarified the exact contrasts driving these omnibus patterns. In evaluations of perceived advertising character, branded posts were judged as significantly more advertising-like than influencer-generated content, producing a paired t-test value of -3.16, an adjusted p-value of 0.026, and an effect size Cohen’s dz of -0.58, representing a medium-sized effect confirmed by non-parametric Wilcoxon analysis. When comparing web-based product pages to influencer posts, differences in perceived commercial character reached significance only in parametric testing. For the fixation-based processing depth metric, no individual pairwise comparisons between specific formats retained statistical significance following Holm correction, suggesting a generalized variance across conditions rather than a dominant format divergence. Direct comparisons between branded and influencer posts revealed no significant differences in attractiveness, product relevance, purchase activation, visual engagement, or scanning activity. These findings challenge the common industry assumption that influencer-generated presentations universally outperform direct corporate brand messaging across subjective and physiological metrics.
Exploratory bivariate Spearman correlation analyses conducted on participant-level aggregated scores uncovered important distinctions between visual attention and consumer appraisals. Following Holm adjustments for multiple comparisons, global visual engagement exhibited a statistically significant negative correlation with perceived product relevance (rho = -0.502, adjusted p = 0.042). This inverse relationship indicates that spending greater aggregate time and fixations examining a stimulus was associated with lower perceived relevance, suggesting that intensive visual exploration may reflect consumer confusion, visual clutter, or informational uncertainty rather than genuine product enthusiasm. Within influencer-generated content, however, perceived influencer affinity demonstrated a statistically significant positive association with declared purchase activation (rho = 0.465, 95 percent confidence interval [0.126, 0.707], p = .010), demonstrating that personal trust and relational closeness, rather than the format itself, track purchase willingness. This positive correlation underscores that consumer response to influencer marketing hinges on perceived relational credibility rather than mechanical format adoption.
Targeted linear mixed-effects models incorporating random participant intercepts and fixed effects for product and content type confirmed that content type remained significant for perceived advertising character (chi-square = 13.78, p = .001) and fixation-based processing (chi-square = 6.82, p = .033). Nonetheless, the authors urged caution when interpreting the exploratory findings, highlighting notable methodological boundaries. The study utilized a small convenience sample of 30 young Slovak students, which a post hoc sensitivity analysis showed was powered primarily to detect medium-to-large effects (dz = 0.53, |r| = 0.49), leaving smaller effects undetected without formal equivalence testing. Furthermore, visual stimuli naturally varied in layout, text volume, and information density, meaning five-second fixation differences may reflect reading demands rather than pure communication context. Global eye tracking without areas of interest prevented isolating fixations on specific elements like prices or faces, and declared purchase activation reflects self-reported intention rather than verified transactional behavior. Consequently, marketing practitioners must recognize that appearing less promotional does not guarantee commercial success.
Subject of Research: Visual processing and subjective evaluation of digital product presentation across branded, influencer, and web formats
Article Title: Less like an Ad, not necessarily more effective: an exploratory eye-tracking study of subjective evaluation and visual processing of digital product content
Article References: Štefko, R., Loumová, V., & Miško, D. (2026). Less like an Ad, not necessarily more effective: an exploratory eye-tracking study of subjective evaluation and visual processing of digital product content. Management & Marketing, 21(1), Article 18. https://doi.org/10.1007/s44491-026-00019-4
Image Credits: AI Generated
DOI: 10.1007/s44491-026-00019-4
Keywords: eye tracking, influencer marketing, digital advertising, visual attention, fixation duration, branded content, commercial intent, persuasion knowledge, product relevance, purchase activation, consumer behavior, neuromarketing, saccadic movement, post attractiveness, advertising recognition, source credibility, digital marketing, visual processing
Cite Scienmag News
Gavin Prescott. (November 18, 2026). Influencer Posts Seem Less Like Ads Without Increasing Consumer Purchase Interest. Scienmag. https://scienmag.com/influencer-posts-seem-less-like-ads-without-increasing-consumer-purchase-interest/
Gavin Prescott. "Influencer Posts Seem Less Like Ads Without Increasing Consumer Purchase Interest." Scienmag, 18 November 2026, https://scienmag.com/influencer-posts-seem-less-like-ads-without-increasing-consumer-purchase-interest/. Accessed 18 November 2026.
Gavin Prescott. "Influencer Posts Seem Less Like Ads Without Increasing Consumer Purchase Interest." Scienmag. November 18, 2026. https://scienmag.com/influencer-posts-seem-less-like-ads-without-increasing-consumer-purchase-interest/








