Influencing Beauty Perceptions: Role of TikTok Influencer Information Adoption in Shaping Consumer Views of Cosmetic Product Quality

Abstract

This research examines how influencer information spreads and is accepted by consumers, focusing on a Tunisian sample of social media users, and how these effects percolate into consumers’ perception of the quality of cosmetic products. Drawing on the Information Adoption Model (IAM), this study develops a conceptual framework adapted to the social media landscape, particularly the TikTok platform. To test this framework, we conducted a survey targeting 285 consumers using a non-random sampling frame, primarily through Facebook and Instagram. The findings suggest that consumers perceive influencer information as useful when they believe it is credible and of high quality. Interestingly, while high-quality information tends to lead to influencer information adoption, credibility alone does not guarantee adoption. Additionally, our study emphasizes the role of influencer information usefulness in driving its adoption. One notable discovery is the link between influencer information adoption and consumers’ perceptions of the quality of cosmetic products. However, this correlation does not hold equally for both genders, thus suggesting a moderation effect between gender and influencer information processing in this context.

1. Introduction

Due to the influence of social media, online shopping has experienced a surge, resulting in a significant increase in internet purchases (McClure & Seock, 2020). Social media reviews have a significant impact on buying decisions (Indrawati et al., 2023). Customer feedback posted on social media can significantly impact a brand’s reputation, both positively and negatively (Rust et al., 2021). Amidst the shifting landscape of social media, TikTok has emerged as a significant player. Its focus on video content and growing popularity have made it a prime platform for content marketing and influencer marketing. This has prompted brands to adapt their strategies to engage with audiences effectively. TikTok’s impact is particularly notable in the beauty industry, with its large user base and high engagement with beauty-related content highlighting the platform’s importance. The global beauty and personal care market is estimated to reach USD$677.19 billion by 2025, growing at a compound annual growth rate (CAGR) of 3.37% between 2025 and 2030 (Statista, 2025). TikTok has over 1 billion active users worldwide as of 2023, demonstrating rapid growth. Beauty-related content consumption on TikTok has increased by 38% in the past year. (Beauty Independent, n.d.). This study investigates how TikTok recommendations and influencer information shape consumers’ perceptions of beauty brands. Prior research in this area is limited. We utilize the Information Adoption Model (IAM) framework to investigate how consumers incorporate information from TikTok influencers into their assessments of product quality. This research contributes to the understanding of the impact of social media platforms on consumer behavior and the perception of quality in the beauty brand sector. This study contributes theoretically by highlighting the connection between influencer information adoption and the perceived quality of cosmetics shared on TikTok and by examining how perceptions of cosmetic product quality are influenced by gender.

Therefore, through this study, we answered the following research questions:

(1)

How does influencer information adoption on TikTok affect the perceived quality of cosmetic products by online consumers?

(2)

What is the moderating role of gender in the relationship between influencer information adoption and perceived quality of cosmetic products?

This study provides practical insights for beauty brands to optimize influencer marketing on TikTok, enhancing consumer perceptions of product quality. Theoretical implications include advancing our understanding of electronic word-of-mouth (e-WOM) through the application of the IAM to TikTok influencer marketing, thereby shaping our understanding of the influence of digital platforms on consumer behavior. Overall, the study serves to bridge theory and practice, offering actionable recommendations for brands while contributing to the academic discourse on influencer marketing and consumer behavior research.

The structure of the remainder of this paper is as follows. First, an in-depth review of the existing literature on information adoption lays the groundwork for our study. Second, the paper will present the research model and the development of hypotheses. Third, the study discusses research methodology, data analysis, and results. A discussion of the theoretical and practical implications derived from the study will follow. Finally, the paper concludes by highlighting the study’s limitations and offering helpful guidance for future research.

2. Literature Review

The rise of social media platforms has significantly altered consumer behavior, particularly in terms of electronic word-of-mouth and influencer marketing (Mrisha & Xixiang, 2024). Researchers have been particularly interested in how content created by influencers impacts consumers’ thoughts and purchasing decisions (Horrich et al., 2024a; Babić Rosario et al., 2020). This interest is fueled by the rapid growth of social media platforms like TikTok, which have become hubs of content creation and sharing. Numerous studies have highlighted the impact of influencer marketing on consumers’ opinions and purchasing behaviors. Influencers, who are often regarded as experts in their fields, utilize product endorsements and recommendations to influence their followers’ purchasing decisions. To effectively market and study influencer marketing, it is essential to understand how consumers receive and accept influencer information. Electronic word-of-mouth (eWOM) plays a crucial role in understanding how information spreads online (Cheung et al., 2008). It involves consumers sharing their thoughts, advice, and experiences in digital spaces (Sulthana & Vasantha, 2019). eWOM greatly influences how people view brands and products, and it can have a stronger impact on purchasing decisions than conventional advertising (Babić Rosario et al., 2020).

The Information Adoption Model (IAM) is widely used to explain how individuals evaluate and integrate information in online environments (Sussman & Siegal, 2003; Erkan & Evans, 2018). Building on the Technology Acceptance Model (TAM) and the Elaboration Likelihood Model (ELM), IAM considers key factors such as argument quality, source credibility, and perceived usefulness to understand how information influences consumer decision-making, particularly in the context of influencer marketing. Despite extensive research on influencer marketing and electronic word-of-mouth (eWOM), there is a limited understanding of how platforms like TikTok influence consumer perceptions of specific product categories, such as beauty products. Although studies have explored the general effects of influencer marketing on consumer behavior, a gap remains in research examining the role of TikTok influencers in shaping consumer perceptions of beauty brands and products.

3. Theoretical Framework: The Information Adoption Model (IAM)

The spread of influencer information, through electronic word-of-mouth (eWOM) conversations, enables consumers to share and exchange ideas (Babić Rosario et al., 2020). However, it is interesting to note that the same content can be interpreted differently by individuals, which emphatically shows the extent to which influencer information can have a differentiated impact (Babić Rosario et al., 2020). Researchers have explored how recipients internalize influencer information, focusing on the process of information adoption. The Information Adoption Model (IAM), developed by Sussman and Siegal (2003), provides a framework for understanding how individuals adopt and integrate information into their behavior and intentions on digital platforms. It combines elements from the Technology Acceptance Model (TAM), the Theory of Reasoned Action (TRA), and the Elaboration Likelihood Model (ELM) to explain how users process and respond to information in computer-mediated communication environments (Erkan & Evans, 2018). Previous studies (e.g., Lee & Workman, 2020) have highlighted that source credibility and entertainment cues play varying roles across platforms. Gendered responses to social media influence have also been explored (Lee & Workman, 2020), highlighting differences in emotional and cognitive processing that may influence adoption. This model consists of four components: (1) argument quality (central route); (2) source credibility (peripheral route); (3) information usefulness; and (4) information adoption (Sussman & Siegal, 2003). The ELM focuses on how audiences evaluate messages, and the theory suggests that persuasion occurs through both central and peripheral routes (Petty & Cacioppo, 1986). On the one hand, the central route, similar to the argument quality component of the IAM argument, suggests that motivated individuals deeply engage with information by considering factors such as its quality when making decisions (Petty et al., 1983). On the other hand, the peripheral route, similar to source credibility in IAM, indicates that motivated individuals rely on heuristics like source credibility or the volume of information for decision-making (Bueno & Gallego, 2021). However, some critics have raised concerns about the focus of AIM on information characteristics in studies related to electronic word of mouth (eWOM). To address this criticism, Erkan and Evans (2018) propose enhancing IAM by incorporating the Theory of Reasoned Action TRA components that consider consumer behaviors towards information. This enhancement aims to identify the factors influencing eWOM in social media platforms that impact consumer purchase intentions.

In the world of electronic word of mouth (eWOM), the significance of having trustworthy information is crucial (Sulthana & Vasantha, 2019). Furthermore, when people perceive information as valuable, they are more likely to take action based on it. This concept is supported by Sussman and Siegal (2003). The IAM has been extensively used in eWOM research due to its focus on understanding the influence of information in computer-mediated communication. Researchers such as Cheung et al. (2008) have utilized IAM in discussion forums, while Shu and Scott (2014) have applied it in media contexts. Considering this study’s emphasis on eWOM within media platforms like TikTok, the IAMs application becomes highly relevant. As shown in Figure 1, we consider IAM components, such as information quality, credibility, usefulness, and adoption, as elements for analyzing how TikTok influences consumers’ perceptions of products.

4. Conceptual Framework and Hypotheses Development

4.1. Relationship Between Information Quality and Information Usefulness

In the field of marketing, it is widely recognized that the fundamental quality of information plays a key role in influencing customer decisions. In shopping scenarios, customers’ willingness to embrace and utilize information for their purchase decisions greatly depends on how they perceive the usefulness of that information (Leong et al., 2021; Fu et al., 2020). Within communities where people frequently share ideas and opinions about products or services, individuals create their judgments about the usefulness of these opinions to guide their informed purchasing decisions (Willemsen et al., 2011). Consequently, if community members find an online comment valuable, they are more likely to incorporate those arguments into their decision-making process. Research has consistently shown that perception of opinion utility predicts the intention to adopt an idea (Petty & Cacioppo, 1986). Information quality, which encompasses persuasiveness and utility, depends on factors such as content relevance, accuracy, presentation style, and timeliness (Bhattacherjee & Sanford, 2006). Prior research, such as Erkan and Evans (2018) and Cheung et al. (2008), has already explored how information quality and source credibility shape consumer attitudes in digital environments, providing foundational insights for this study. For instance, a review that includes photos, a description, and a convincing commentary tends to be carefully evaluated and embraced (Horrich et al., 2024b; Kozinets et al., 2010). In the realm of communication, when it comes to products, services, and guidance on purchases, having valuable information can greatly contribute to customers’ understanding of a product’s usefulness (Park et al., 2007). The presented arguments can play a role in helping consumers make informed decisions about the worth of a particular product or service.

Moreover, as stated by Sussman and Siegal (2003), the IAM suggests that the quality of information, in communications, plays a role in determining its usefulness. Therefore, the strength of the arguments presented has an impact on how we evaluate the usefulness of a product or service, aligning with IAM principles. Székely (2020) has emphasized the nature of information quality, emphasizing accuracy, completeness, and timeliness as aspects that meet users’ expectations. Supporting this view, Park et al. (2007) found a relationship between information quality and purchase intention. Similarly, Erkan and Evans (2018) have highlighted that information quality influences its utility. As a result, we predict that there is, indeed, an association between information quality and its utility. Furthermore, Sussman and Siegal (2003) noted that when users perceive information as valuable or useful, they are more likely to adopt it. Thus, information quality could serve as a predictor of its utility. Based on these insights, we propose the following hypothesis:

H1. 

Information quality about cosmetic products on TikTok positively influences information usefulness.

4.2. Relationship Between Information Credibility and Information Usefulness

In the field of marketing, researchers such as Petty and Cacioppo (1986) have emphasized the role that persuasive information plays in influencing user decision-making. According to McKnight and Kacmar (2007), individuals tend to trust the information presented to them by influencers. Similarly, Smith and Vogt (1995) suggest that readers’ assessment of information reliability reflects their level of agreement with a viewpoint. As a result, readers tend to adopt information that they consider trustworthy (AlBar & Hoque, 2019).

In contexts where consumers obtain product and service information, as well as purchasing advice, from fellow consumers on platforms and communities, source credibility becomes crucial in assessing the usefulness of such information (Wu & Wang, 2011). Messages and comments about products from individuals are often perceived as trustworthy and influential by consumers when assessing the usefulness of a product or service (Filieri et al., 2018).

Furthermore, when potential customers receive multiple messages from influencers, the trustworthiness of the information becomes extremely important (Filieri, 2016). According to Wallace et al. (2020) and Wathen and Burkell (2002), the credibility of the information is a key factor in the persuasion process, closely associated with the source. Information from credible sources holds great value and helps in spreading information, forming a basis for personal persuasion (Erkan & Evans, 2018). Previous studies have found a connection between information credibility and customers’ intention to make purchases, especially when they perceive the information as valuable and adaptable. Information credibility thus plays a significant role in consumer decision-making (Ahmad et al., 2020). In line with the IAM, the credibility of sources providing messages and comments serves as a significant indicator that heavily influences the usefulness of that information (Sussman & Siegal, 2003). Accordingly, drawing on the IAM principles, we propose that source credibility influences the evaluation of a product’s usefulness. This understanding leads us to formulate the following hypothesis:

H2. 

Information credibility about cosmetic products on TikTok positively influences information usefulness.

4.3. Relationship Between Information Quality and Information Adoption

The concept of information quality, as discussed by Wang and Yan (2022), centers on the impact of the words used to convey the essence and meaning of a message. In today’s era, where the Internet is widely accessible, individuals from any location can freely express their opinions and reviews about various products or services (Ali et al., 2021). The democratization of information sharing presents challenges for consumers in evaluating the quality of information (Puaschunder, 2021). Cheung et al. (2008) discovered that the relevance and completeness of information significantly influence its acceptance. High-quality information can notably assist consumers in accessing timely content (Cheung et al., 2008). Moreover, Moore and Lafreniere (2020) highlighted the pivotal role of emotional cues in electronic word-of-mouth (eWOM), where expressions such as happiness or anger can clarify the sender’s viewpoint and enhance the diagnostic effect of the information shared. Pan and Zhang (2011) also proposed that comment length could be an indicator of their usefulness, as longer comments provide more details but require cognitive effort to process, which may make them seem and actually be more persuasive (Moore & Lafreniere, 2020). According to Park et al. (2007), the quality of reviews for products and services plays a significant role in influencing consumers’ decision-making processes and their intentions to purchase. When reviews are clear, substantial, and reliable, they provide consumers with accurate information to evaluate products or services, thereby carrying weight in their decision-making process (Mudambi & Schuff, 2010). Building on this idea, Huang et al. (2011) conducted a study on the impact of electronic word of mouth (eWOM) and identified information quality as a significant factor that positively affects consumers’ adoption and acceptance of information. Based on the above, several cues may indicate that the information consumers access from other consumers is of high quality, which in turn may facilitate their adoption of that information. Based on these findings, we propose the following hypothesis:

H3. 

Information quality about cosmetic products on TikTok positively influences information adoption.

4.4. Relationship Between Information Credibility and Information Adoption

The reliability and trustworthiness of information from a source define information credibility as perceived by consumers (Kang & Namkung, 2019). Since perception is subjective, the assessment of information credibility often varies among consumers, introducing an element of uncertainty. Awad and Ragowsky (2008) emphasize that information credibility plays a crucial role in the purchasing decision-making process. Consistently, several studies have highlighted a connection between consumer purchase intentions and the credibility of information (Martínez et al., 2020), underscoring its significant impact on consumer decisions, particularly regarding purchase intentions.

Zhang and Watts (2008) noted that information perceived as having source credibility is more likely to be adopted. In virtual community environments, consumers assess the credibility of sources based on characteristics such as identity exposure, popularity, community status, and more (Chaiken, 1980). In line with this idea, B. L. Song et al. (2021) suggested that information credibility is primarily determined by how consumers perceive the quality of the information and the level of trust it earns from a reputation standpoint. According to studies conducted by Arenas-Márquez et al. (2021) and Forman et al. (2008), the influence of popular individuals has a significant impact on the effectiveness of word-of-mouth marketing in online communities. These influencers often possess a wealth of knowledge, respect and innovative ideas, making them influential within their communities. The quality of information plays a role in helping users assess its reliability. This, in turn, influences consumers’ decision-making process and their likelihood of adopting the information presented to them for making purchase decisions (Forman et al., 2008). Previous research conducted by Luo et al. (2015) has confirmed that the quality of arguments put forth in reviews positively affects their credibility. Consequently, consumers are more likely to adopt information from a platform if they find it credible and trustworthy during their decision-making process (Luo et al., 2015). Building upon these findings, Cheung et al. (2008) discovered that consumers’ perception of information credibility is a factor influencing their adoption of information. Based on these insights and previous studies, we propose the following hypothesis:

H4. 

Information credibility about cosmetic products on TikTok positively influences information adoption.

4.5. Relationship Between Information Usefulness and Information Adoption

According to Hussain et al. (2020), useful information provides helpful, educational, and valuable arguments and indicators. When information is helpful, it is more likely to be adopted and used effectively (Erkan & Evans, 2018). Helpful information is instructive and provides insights that can support people’s beliefs about how to improve their performance (Cheung et al., 2008). When people perceive information as beneficial, they are more likely to use it. They also tend to seek out information that aligns with their needs and goals (Sardar et al., 2021). When consumers perceive the information they acquire as useful and advantageous in their decision-making process, this perception translates into the perceived usefulness of the information itself, which subsequently affects its adoption (Sussman & Siegal, 2003). Several researchers have identified information usefulness as a predictor of both information adoption and purchase intention (Erkan & Evans, 2018; Sussman & Siegal, 2003). Filieri (2015) noted that the primary factor influencing information adoption behavior is how beneficial the information is perceived to be in diagnosing a specific situation. Previous studies have also discovered a relationship between the acceptance of information and its practicality (Tien et al., 2019). Moreover, the impact of information on the receiver largely depends on its reliability and value. Therefore, if information is not considered reliable or valuable by the recipients, it is unlikely to have any effect on them (Hussain et al., 2017). Chu and Kim (2011) found that when social media users perceive a large amount of information, they tend to engage in extensive electronic word-of-mouth (eWOM) discussions and exhibit a greater intention to adopt new information (S. Song et al., 2021; Erkan & Evans, 2018). They emphasized that in situations where consumers gather information from others who have made purchases, there is evidence of adopting useful information, indicating a positive relationship between information usefulness and purchase intention. In light of these insights, this study proposes the following hypothesis:

H5. 

Information usefulness about cosmetic products on TikTok positively influences the information adoption.

4.6. Relationship Between Information Adoption and Perceived Product Quality

The primary objective of this study is to examine how the adoption of information influences the perceived quality of beauty products. The underlying premise is that when users are satisfied with online-generated information, it enhances their perception of product quality, creating a sense of comfort and security that influences their purchasing decisions (Mofokeng, 2021). Thus, according to the user’s perspective, higher levels of information quality and its subsequent adoption lead to an increased perception of product quality (Mofokeng, 2021). Furthermore, the role of gender is significant in shaping consumer behavior, particularly in the relationship between consumers and brands, as well as how customers perceive products (Lee & Workman, 2020). Men and women have different perspectives, opinions, and values, which shape their unique needs and preferences regarding products and services (Okazaki & Mendez, 2013a, 2013b). Gender is now a significant factor in marketing, particularly with the widespread use of social media. It affects how consumers respond to marketing messages and makes gender a key consideration in segmenting consumers and designing marketing campaigns. Marketers utilize this understanding of gender differences to tailor their strategies and communications, thereby better appealing to different groups of consumers.

Based on the discussion points, this study puts forward the following hypotheses:

H6. 

The adoption of information about cosmetic products on TikTok positively influences the perceived quality of these products.

H7. 

The adoption of information about cosmetic products on TikTok positively influences the perceived quality of these products, with gender acting as a moderating factor.

Figure 2 below presents the IAM, which outlines the theoretical structure guiding our study. As shown, the model incorporates argument quality and source credibility as antecedents of perceived usefulness and adoption.

5. Methodology

5.1. Sample and Data Collection

To practically test the conceptual model, we decided to gather evidence using a quantitative research approach. We conducted a survey designed to capture a range of responses. To ensure the reliability and validity of this study, we employed multiple-item scales. The survey consisted of 18 statements, and participants were asked to indicate their level of agreement or disagreement with each statement on a five-point Likert scale. The scale ranged from “1” (strongly disagree) to “5” (strongly agree), with “3” (neutral) as the midpoint. To specify the sample size, we followed a principle elaborated by Jöreskog and Sörbom (1982), which suggests having at least 10 respondents per survey item. This approach resulted in a total of 285 responses. The final sample consisted of 285 individuals who actively use Facebook and Instagram.

Our study encompassed a diverse array of Facebook and Instagram users, strategically selected to create a representative pool of individuals with varying levels of exposure to social media content, including the TikTok platform. To ensure the relevance of responses, a filter question was placed at the beginning of the questionnaire. Only individuals who confirmed they were active TikTok users were allowed to proceed with the survey. By focusing on users of these widely used platforms, we aim to identify the overarching patterns in digital marketing and consumer behavior. We took great care to ensure the relevance of our sample by specifically inquiring about TikTok usage, allowing us to glean valuable insights from those with first-hand encounters of beauty-related content on the platform. Ultimately, by distributing the questionnaire on both Facebook and Instagram, we were able to provide comprehensive insights into the impact of social media on consumer perceptions of beauty products, bolstering our primary focus on TikTok. We ensured a balance in gender representation in our sample, with 49.1% male participants and 50.9% female participants. All age categories were represented, with 5.6% of the respondents being under 18 years old; 60.4% falling between the ages of 18 and 30; 25.3% falling between the ages of 31 and 40; 6% between the ages of 41 and 49; and 2.8% being 50 and over. Moreover, in terms of usage by respondents, there was also a diversity in how participants used the platform; 24.6% used TikTok for two hours per day; 22.1% used TikTok for an hour every day; 28.1% used it for less than an hour per day; and 25.3% used it for an hour each week. This indicates that our sample encompasses a diverse range of user profiles.

The survey was initially developed in English and then translated into Arabic to ensure accessibility for the Tunisian population. Bilingual experts conducted a back-translation process to ensure linguistic equivalence. The questionnaire was pre-tested with 20 respondents to assess clarity, coherence, and relevance. To address potential common-method bias, we conducted Harman’s single-factor test, which revealed no dominant single factor, indicating that common-method variance was not a serious concern.

5.2. Measurement Scales

In conducting this study, we utilized the past literature to select and adapt the measurement scales. The measurement items used in this study are listed in Appendix A. These scales have been extensively used in marketing research, and as such, they are recognized for their robust psychometric properties, ensuring the accuracy and credibility of our data. The scale for assessing information quality consisted of four items and was adapted from Erkan and Evans (2018) and Park et al. (2007). To measure information credibility, a four-item scale was employed, drawing from the scales developed by Erkan and Evans (2018). The construct of information usefulness was measured using a four-item scale adapted from Hussain et al. (2020). This scale is designed to assess the extent to which users find the information helpful and beneficial. To operationalize the concept of information adoption, we utilized the three-item scale developed by Shen et al. (2014). This scale focuses on the extent of user engagement with and utilization of the information. The perceived product quality was measured using a three-item scale derived from Washburn and Plank (2002). This scale is designed to evaluate the overall quality perception of the products by the consumers.

5.3. Data Analysis

In this study, we conducted data analysis on the final sample of 285 users who play a role in deciding which cosmetic products to purchase. To begin our analysis, an exploratory factor analysis (EFA) was conducted using SPSS v30 to assess the underlying structure of the measurement items and ensure construct validity before proceeding with the structural model analysis. For our focused analysis and hypothesis testing, we utilized SMARTPLS 4 and employed the Partial Least Squares (PLS) method. This two-step approach, combining SPSS for data analysis and SmartPLS for structural equation modeling, allowed us to thoroughly examine the data in line with the study’s goals and hypotheses.

6. Results

6.1. Assessment of the Measurement Model

During our research, we thoroughly examined the reliability of our measurement model through a comprehensive analysis of both convergent and discriminant validity. To assess convergent validity, we meticulously reviewed factor loadings, composite reliability (CR), and average variance extracted (AVE). As shown in

Table 1. Average variance extracted (AVE).

Construct	AVE	Composite Reliability	Cronbach’s Alpha
Information adoption	0.779	0.914	0.859
Information credibility	0.757	0.926	0.893
Gender
Information quality	0.677	0.893	0.840
Perceived product quality	0.792	0.920	0.869
Information usefulness	0.757	0.926	0.893

. The CR metric, which measures the reliability and internal consistency of a measurement scale, yielded strong Cronbach’s alpha values above 0.7 for all scales, indicating a high level of consistency. Moreover, our AVE values were all above 0.50, suggesting that a majority of the variability in each construct could be attributed to its associated items. To confirm discriminant validity, we performed a rigorous loading analysis to ensure that the indicator loadings for each construct were higher than any cross-loadings on other constructs. Table 1 presents the results of the measurement model evaluation, including AVE, composite reliability, and Cronbach’s alpha for each construct, while

Table 2. Heterotrait–monotrait ratio (HTMT).

	Information Adoption	Credibility of the Information	Gender	Quality of Information	Perceived Quality	Usefulness of the Information	Gender × Adoption of Information
Information Adoption
Credibility of the information	0.682
Gender	0.513	0.585
Quality of information	0.751	0.785	0.507
Perceived quality	0.717	0.818	0.809	0.701
Usefulness of the information	0.827	0.802	0.456	0.811	0.710
Gender × Adoption of Information	0.834	0.421	0.357	0.532	0.540	0.608

shows the heterotrait–monotrait (HTMT) values.

6.2. Assessment of the Structural Model

When evaluating our model, we considered two key factors: the coefficient of determination (R²) and predictive relevance (Q²). The R² values for variables such as ‘information adoption’ (0.564), ‘perceived quality’ (0.662), and ‘information usefulness’ (0.606). Our data analysis revealed that all Q² values exceeded 0.1, indicating that our model has exceptional predictive accuracy and reliability (Hair et al., 2014).

Table 3. Explanatory power of the model.

Construct	R-Square
Information adoption	0.564
Perceived product quality	0.662
Information usefulness	0.606

shows the explained variance (R²) of each dependent variable, indicating strong predictive power for information adoption, usefulness, and perceived product quality.

Moreover, the model’s predictive relevance is supported by the fact that Q² values surpassing 0.1 demonstrate predictive relevance. The Q² results presented in Table 3, all of which exceed the 0.1 threshold, confirm that our model is statistically reliable. We also examined multicollinearity using Variance Inflation Factors (VIF). All VIF values were below 3.3, indicating no critical multicollinearity between constructs such as credibility and usefulness.

Table 4. Predictive relevance.

	Q2 Predict
Information adoption	0.449
Perceived product quality	0.369
Information usefulness	0.597

presents the Q² predictive relevance values, which confirm the model’s robustness.

For the overall unified structural model, the fit indices (SRMR = 0.055, NFI = 0.91, CFI = 0.94) indicate an acceptable model fit. These values reflect the model, which includes all variables and hypotheses tested simultaneously (Hair et al., 2014).

The structural model was estimated in three parts to reflect the sequence of relationships between variables: (1) predictors of information usefulness, (2) predictors of information adoption, and (3) predictors of perceived product quality. Each sub-model builds upon the previous one. However, we also tested a complete structural model that incorporates all relationships simultaneously to assess the overall predictive power of the conceptual framework (Figure 2). The results remained consistent with those reported in the individual models, confirming the robustness of the findings. While we report results from a complete structural model, we also present results across three sub-models to enhance clarity and interpretability. Each sub-model addresses a different dependent construct (i.e., perceived usefulness, information adoption, and perceived product quality), enabling a focused analysis of each relational block. This approach aligns with previous studies that have employed complex, multi-stage models (Ertz et al., 2016, 2024). However, the complete hypothesis testing and model-fit indices are also reported for the overall model to ensure transparency and facilitate comparison.

6.3. Hypothesis Testing

During the analysis of our hypotheses, we utilized statistical indicators, including p-values and t-values. To establish significance, we set the threshold for a t-statistic value above 1.96 and a p-value below 0.05. The findings, detailed in Table 4, indicate that all our hypotheses received support, confirming their relevance and validity within our research framework.

Table 5. Hypotheses testing.

Hypothesis.		Original Sample (O)	Standard Deviation	t-Statistics	p-Value	Decision
1	Information quality → Information usefulness of	0.412	0.059	7.025	0.000	Accept
2	Information credibility of → Information usefulness	0.437	0.063	6.887	0.000	Accept
3	Information quality → Information adoption	0.221	0.080	2.760	0.005	Accept
4	Information credibility → Information adoption	0.097	0.094	3.035	0.001	Accept
5	Information usefulness → Information adoption	0.500	0.081	6.141	0.000	Accept
6	Information adoption → Perceived product quality	0.285	0.090	3.159	0.002	Accept
7	Gender × Information adoption → Perceived product quality	0.106	0.108	2.988	0.003	Accept

summarizes the hypothesis testing results, including standardized path coefficients, t-values, and significance levels.

Overall, the results support all the proposed hypotheses. Information quality and credibility both positively influence perceived usefulness, with information quality showing a stronger effect. Perceived usefulness, in turn, significantly predicts information adoption, which has a substantial impact on perceived product quality. The effect of credibility on information adoption is statistically significant. The moderation analysis reveals that gender has a significant influence on the relationship between information adoption and perceived product quality. These results confirm the relevance of the IAM framework in explaining how influencer content shapes consumer perceptions in a short-form video context. These findings show that information quality has the strongest influence on perceived usefulness, aligning with the importance of clarity and relevance in social media content. Interestingly, source credibility, though significant, had a lower impact, suggesting that users may prioritize content quality over influencer status. The gender-based moderation indicates deeper behavioral variance in content adoption, especially among women, possibly linked to emotional engagement patterns noted in the previous literature.

7. Discussion of the Results

In this study, we delved into the dynamics of information adoption on TikTok and its consequential impact on the perceived quality of cosmetic products. Our empirical findings, primarily anchored in the IAM, unravel several pivotal insights. Firstly, we confirm the significant influence of information quality on its perceived utility (H1), echoing Erkan and Evans (2018). This revelation highlights the importance of delivering detailed and accurate content to TikTok’s discerning users in the cosmetics industry. The observation further complements this result that information credibility substantially enhances perceived utility (H2), aligning with the assertions of Sussman and Siegal (2003), and underlining the indispensable role of credibility in consumer perception.

Our finding that information quality strongly predicts perceived usefulness is consistent with previous studies on e-WOM and influencer marketing (Erkan & Evans, 2018; Cheung et al., 2008). This reinforces the role of clear, relevant, and structured content in shaping consumer perception, especially in short-form video formats like TikTok.

Interestingly, while we identified a relevant impact of information quality on its adoption (H3), resonant with the studies of Huang et al. (2011) and Cheung et al. (2008), Interestingly, although information credibility was found to have a statistically significant effect on information adoption (H4), the effect size was relatively small (β = 0.097). This suggests that while consumers acknowledge credibility, it may not be the primary driver for adopting influence-shared information on platforms like TikTok. Although credibility showed a statistically significant but weak influence on information adoption, this aligns with studies suggesting that younger consumers often rely more on content engagement and entertainment value than traditional credibility cues (Lee & Workman, 2020). This might explain the relatively low effect size in our study.

The emotional, visual, or entertaining aspects of content may outweigh credibility in this context. This nuance provides a more realistic perspective on how credibility operates and encourages further research into the specific mechanisms that moderate its influence. Our results confirm that information usefulness predicts adoption, in line with Erkan and Evans (2018) and Cheung et al. (2008), who highlighted the role of informational value in e-WOM. However, unlike the findings of Sweeney et al. (2008), credibility played a more modest role here, possibly due to platform-specific dynamics or demographic characteristics of the sample.

Our exploration further substantiates a significant linkage between information adoption and perceived product quality (H6), illustrating the reliance of consumers on adopted information for quality judgments. This insight may be particularly valuable for marketers, as it highlights the impact of information dissemination strategies on consumer perception. Notably, the analysis confirmed that gender significantly moderates the relationship between information adoption and perceived product quality (H7). This suggests that men and women interpret and respond to influencer information differently, underscoring the importance of gender-sensitive marketing strategies on platforms like TikTok. The stronger effect of information adoption on product perception among women aligns with past research, which suggests that female consumers are more sensitive to social and emotional cues in influencer content (Lee & Workman, 2020). This gendered response has been particularly noted in beauty and fashion-related domains.

Our findings, although illuminating, must be contextualized within the confines of our research scope, particularly in relation to the distinctive attributes of TikTok and the cultural nuances of our Tunisian sample. These elements may distinctly shape information perception and adoption, warranting further inquiry into the platform’s unique influence across different demographics. In essence, our research offers a multifaceted understanding of TikTok’s role in shaping consumer perceptions in the cosmetic industry. It highlights the paramount importance of information quality and utility while also suggesting a more nuanced understanding of the role of credibility and demographic factors, such as gender, in influencing consumer behavior. This knowledge is indispensable for marketers seeking to navigate the complex landscape of social media-driven consumer engagement, particularly in the rapidly evolving field of digital marketing.

8. Theoretical and Managerial Implications

This research makes a significant contribution to the study of integrated marketing communication in Tunisia, with a particular focus on the impact of TikTok. By examining TikTok’s unique characteristics, we offer new insights into how consumers’ adoption of information influences their behavior, thereby addressing a gap in current research. Our results support earlier research that underlined the significance of information adoption in consumer decision-making. Moreover, we develop a conceptual model that combines current marketing ideas, thereby deepening our understanding of consumer behavior in the digital era. (Fu et al., 2020). Our research contributes valuable insights to academic discussions by presenting a conceptual framework that not only highlights relationships between variables previously explored, but also integrates findings from established marketing and consumer behavior theories (Erkan & Evans, 2018).

This study investigates the direct impact of using information from social media on how people perceive product quality. It also looks at how gender influences this relationship in Tunisia, a developing country. Our research builds upon the work of Lee and Workman (2020), which highlights the importance of understanding the role of gender in consumer behavior across diverse cultures. We develop a model that illustrates how information adoption on TikTok influences consumer behavior in Tunisia, taking into account the social, economic, and cultural factors specific to developing nations. Our findings contribute to the academic literature by providing insights into the particular context of Tunisia Managerial implications.

From a managerial standpoint, our research offers valuable insights for companies integrating TikTok into their digital marketing strategies, with a particular emphasis on the pivotal role of social media influencers. Our study highlights the importance of influencer-generated content in enhancing the utility and adoption of information. In light of those results, digital marketers should focus on the value and relevance of the messages that influencers spread. This is because consumers are more likely to believe and buy into messages from influencers. When companies deliver high-impact information through influencers, they can change how consumers think and behave. Our research indicates that this can lead consumers to perceive the product as better.

Although we did not test the impact of perceived product quality on actual purchase decisions or buying intentions, it is fair to posit that enhanced information adoption due to better information quality might lead to increased purchases (Mofokeng, 2021).

This research also highlights that the information utility, not just its adoption, is positively influenced by its credibility. Given the potential impact on purchasing intentions and decisions, industry professionals should prioritize the credibility of information to facilitate more effective adoption.

Furthermore, the positive impact of information utility on its adoption suggests that marketers should concentrate on disseminating consumer-relevant information to ensure its adoption and eventual conversion into a purchasing action.

These findings have several practical implications for different stakeholders. Marketers and brand managers should prioritize producing influencer content with high information quality and perceived usefulness to improve product perception. Given the small but significant role of credibility, influencers must maintain authenticity and consistent messaging to build trust over time. The gender moderation effect suggests that targeted strategies may be more effective, with more emotionally resonant or detailed content aimed at female audiences. Platform managers can enhance user engagement by promoting credible influencers and optimizing content formats to improve perceived usefulness, particularly in markets with growing digital influence, such as Tunisia.

Finally, the study reveals that the adoption of information has a favorable effect on the customer’s perception of product quality. Marketers, therefore, should collaborate and disseminate information that, if adopted by customers, can positively influence their perception of a product’s quality, potentially leading to a purchase decision.

9. Limitations and Future Research Avenues

Like any scientific inquiry, this study is subject to certain limitations that warrant consideration. Firstly, the research was conducted exclusively within the Tunisian context with a sample size of 285 respondents. Consequently, extrapolating these results to other countries is challenging due to variations in information and communication technology levels, cultural differences, and customer experiences. Future research should involve comparative studies in diverse cultural settings to refine our tested model further and validate its reliability and generalizability.

Secondly, since this study was conducted explicitly on TikTok, its findings may not be directly applicable to other social media platforms, such as Instagram, Facebook, YouTube, or Vimeo. Future studies should aim to test our model across various social media platforms and in different sectors beyond cosmetics, such as online services, clothing, and electronic devices, to ascertain the model’s versatility and applicability.

Furthermore, it was important that this research did not focus on customers of a specific brand or a particular category of cosmetic products. Future research should thus aim to apply this theoretical model to customers of well-defined brands who purchase particular product categories. Future research could consider including prospects in addition to actual customers for a more comprehensive understanding.

Another limitation concerns the recruitment platform. The questionnaire was distributed via Facebook and Instagram, which could introduce a platform-based sampling bias. To counterbalance that issue, a filter question was used to include only respondents who self-identified as active TikTok users. Nevertheless, we acknowledge that this approach may not fully replicate the immersive and algorithm-driven environment of TikTok, and some aspects of user experience unique to the platform may not have been fully captured. This limitation should be considered when interpreting the results and generalizing the findings. As a result, the generalizability of the findings to the broader TikTok user base should be interpreted with caution. Future research should aim to sample active TikTok users for a more platform-specific perspective directly. Although gender moderation was tested using an interaction effect approach, future research could employ multi-group analysis (MGA) to further examine model differences between male and female participants by comparing model fit and path strength across groups.

Finally, future research should investigate the indirect relationships within the IAM and perceived product quality to identify any potential indirect impacts. Additionally, examining other moderating variables such as generational differences could provide deeper insights, as this study primarily focused on direct relationships between variables.

In summary, while our study provides valuable insights into information adoption on TikTok and its impact on perceived quality in the cosmetics sector, there are ample opportunities for future research to expand on these findings, enhancing the model’s robustness and its applicability across different contexts and platforms. Future studies could also investigate the role of brand-related variables, such as brand value, brand familiarity, and brand awareness, in shaping the effectiveness of influencer content. These elements may moderate how consumers process and adopt information shared by influencers, particularly in competitive product categories such as beauty and personal care.