Many marketers and businesses have used social media influencers to enhance their brand awareness. These influencers act as endorsers who link brands and consumers, fostering consumer engagement. This paper aims to systematically review the live-stream influencer marketing literature to identify key theoretical models and influential factors from the user (influencers and consumers) perspective. Additionally, we also provide insights for future research directions.
We adopted a systematic literature review (SLR) approach to synthesise research unbiasedly and comprehensively. In the preliminary stage, 1,144 articles were found; after thorough screening, 93 research papers were selected for review and synthesis.
The findings from this SLR allow us to better understand the application of live-stream influencer marketing theories to consumer purchase behaviour. We identified the widely used theories, such as the stimulus-organism-response model, along with key motivational factors influencing viewers (potential consumers), their attitudes and behaviours (including purchase intention), and other related aspects.
This SLR is among the first to investigate the theories used and motivational factors of live-stream influencer marketing on consumer purchase intention.
1. Introduction
Marketers have increasingly used social media to build social presence (Cabeza-Ramírez et al., 2022; Yin et al., 2025) and generate customer value (Bu et al., 2022). Social media influencers, also referred to as online opinion leaders, create content that is widely disseminated and acknowledged by a broad audience (Zhang et al., 2023b). Consequently, businesses are allocating larger portions of their budgets to influencer marketing—especially through live-streaming—as a strategic approach to product promotion (Long et al., 2024a; Zhang et al., 2023b). The rising adoption of live-streaming features on social media and e-commerce platforms is evident, with live-stream sales reaching $5,626.2 billion in 2023 and projections of increasing to $8,439.3 billion by 2025 (Fan et al., 2024).
Scholarly interest in live-stream influencer marketing has also expanded rapidly, with studies highlighting the substantial impact of influencer endorsements on sales performance (Zhang et al., 2024a; Meng and Lin, 2023; Rungruangjit, 2022). These endorsements influence content appeal, consumer attention, and purchase intentions (Yang et al., 2024a; Cui et al., 2024). However, despite the extensive research on conventional influencer marketing, a comprehensive understanding of the motivational factors and behaviour effects specific to live-stream influencer marketing remains lacking (Fletcher and Gbadamosi, 2024). Unlike conventional influencer marketing, live-streaming fosters real-time interactions and social presence, which may potentially influence consumer trust, engagement, and purchase decisions in distinct ways.
Classifying and defining motivational forces remains a core challenge in explaining consumer behaviour within live-stream marketing. Building on the intrinsic–extrinsic distinction, prior research highlights hedonic and utilitarian values as well as normative social influence as the primary drivers of viewing and purchase behaviour in this context (Cinar et al., 2011; Kim and Drumwright, 2016). In contrast, variables such as product involvement, self-congruence with the streamer, the influencer’s perceived trustworthiness and stable personality traits function mainly as antecedents or moderators that determine how strongly these motives are activated. Once triggered, these motivations feed into downstream processes—namely perceived value, social identity formation, and emotional connection with the influencer—that ultimately shape real-time engagement and purchase outcomes (Confente et al., 2020; Shan et al., 2020). A refined conceptualisation that explicitly differentiates antecedents, motivational forces, mediators, and outcomes is therefore essential for unpacking the complex interplay between consumer motivation and live-stream engagement.
To address this research gap, a systematic literature review (SLR) is necessary to synthesise existing findings, identify theoretical frameworks, and uncover unexplored areas in live-stream influencer marketing. While previous studies examine different aspects of consumer motivation and influence effectiveness, no systematic effort has been made to integrate these insights and assess their collective implications. By adopting an SLR approach, this study aims to first clarify the scope of motivational factors in live-stream influencer marketing, and second, to identify theoretical frameworks that explain consumer behaviour in live-stream marketing. Finally, the study strives to uncover gaps in understanding how interactive engagement shapes purchase decisions.
More specifically, this paper systematically reviews live-stream influencer marketing on consumers’ purchasing behaviours from two perspectives: (1) influencers and (2) consumers. In this SLR, we first explore the widely applied theories in live-stream influencer marketing research, followed by an analysis of the motivational factors and the subsequent impacts on consumer behaviours, particularly in relation to purchase intention. While live-streaming represents a relatively new frontier within social media studies, most existing reviews have predominantly focused on conventional influencer marketing (e.g. Kanaveedu and Kalapurackal, 2024). This review provides comprehensive insights to explore how live-streaming creates different forms of consumer value and how these values influence purchasing behaviour.
We conducted this SLR to build on existing studies of the consumer decision-making process in online platforms and guide future research. Future work should explore factors beyond motivation, such as social presence, to provide a comprehensive view of consumer behaviour. Comparative studies across social media platforms can reveal unique dynamics, while cross-discipline research can examine how diverse values influence engagement. Further, segmentation of online and offline purchase intentions for high-experience products can aid targeted marketing strategies. Finally, emerging technologies offer opportunities to enhance interactivity in live-stream marketing.
The following sections outline our methodology, provide background on live-stream influencer marketing, and review key live-stream influencer marketing theories. We then explore the literature on consumer motivating factors and purchase decision-making, including consumer motivation, attitudes toward endorsements, and sponsorship regulations. We conclude with a discussion, future research directions, and final remarks.
2. Research methodology of literature review
2.1 Search procedure, screening and selection criteria
This paper follows the systematic procedure postulated by Fisch and Block (2018). The search methodology, based on Aljaroodi et al. (2019), was conducted in three distinct stages—plan, conduct and report (Figure 1). First, we identified the need for an SLR, developed a review procedure, and evaluated it. Next, we searched the selected databases, retrieved relevant studies, and further analysed them. Finally, we documented and reported our findings.
The flowchart is divided into three dashed ovals arranged vertically. From top to bottom, the ovals are labeled on the left side as follows: “1) Plan Stage,” “2) Conduct Stage,” and “3) Report Stage.” Inside the “1) Plan Stage” oval, three rectangular boxes are arranged vertically and connected sequentially. The first box is labeled “Step 1.1: Need for S L R,” with a right-pointing arrow leading to a box labeled “Introduction (Section 1).” A downward arrow connects “Step 1.1: Need for S L R” to “Step 1.2: Review Strategy.” “Step 1.2: Review Strategy” is connected on the right to “Search Strategy (Section 2)” and below to “Step 1.3: Evaluate Strategy.” A two-way arrow from “Search Strategy (Section 2)” connects to “Step 1.3: Evaluate Strategy.” “Step 1.3: Evaluate Strategy” connects downward to the “2) Conduct Stage,” oval. In the “2) Conduct Stage,” oval, a box labeled “Web of Science (1013); Scopus (131)” appears on the left above the label of the box. “Step 1.3: Evaluate Strategy,” from the “1) Plan Stage,” oval connects to “Step 2.1: Search (1144).” A two-way arrow from “Search Strategy (Section 2)” also connects to “Step 2.1: Search (1144).” A right arrow from “Step 2.1: Search (1144)” leads to “Abstract or title or keywords (309), which connects downward to “Full-Text Review of Studies.” Below “Step 2.1” is “Step 2.2: Select (93),” which has a left-pointing arrow from “Full-Text Review of Studies.” “Step 2.2: Select (93)” connects downward to “Step 2.3: Extract or Analyse,” which further connects downward to “Step 3: Report Findings,” in the “3) Report Stage,” oval. Finally, “Step 3: Report Findings” points right to a box labeled “Results (Section 3),” completing the flowchart.Stages of our SLR. Source: Authors’ own creation
The flowchart is divided into three dashed ovals arranged vertically. From top to bottom, the ovals are labeled on the left side as follows: “1) Plan Stage,” “2) Conduct Stage,” and “3) Report Stage.” Inside the “1) Plan Stage” oval, three rectangular boxes are arranged vertically and connected sequentially. The first box is labeled “Step 1.1: Need for S L R,” with a right-pointing arrow leading to a box labeled “Introduction (Section 1).” A downward arrow connects “Step 1.1: Need for S L R” to “Step 1.2: Review Strategy.” “Step 1.2: Review Strategy” is connected on the right to “Search Strategy (Section 2)” and below to “Step 1.3: Evaluate Strategy.” A two-way arrow from “Search Strategy (Section 2)” connects to “Step 1.3: Evaluate Strategy.” “Step 1.3: Evaluate Strategy” connects downward to the “2) Conduct Stage,” oval. In the “2) Conduct Stage,” oval, a box labeled “Web of Science (1013); Scopus (131)” appears on the left above the label of the box. “Step 1.3: Evaluate Strategy,” from the “1) Plan Stage,” oval connects to “Step 2.1: Search (1144).” A two-way arrow from “Search Strategy (Section 2)” also connects to “Step 2.1: Search (1144).” A right arrow from “Step 2.1: Search (1144)” leads to “Abstract or title or keywords (309), which connects downward to “Full-Text Review of Studies.” Below “Step 2.1” is “Step 2.2: Select (93),” which has a left-pointing arrow from “Full-Text Review of Studies.” “Step 2.2: Select (93)” connects downward to “Step 2.3: Extract or Analyse,” which further connects downward to “Step 3: Report Findings,” in the “3) Report Stage,” oval. Finally, “Step 3: Report Findings” points right to a box labeled “Results (Section 3),” completing the flowchart.Stages of our SLR. Source: Authors’ own creation
We selected the Web of Science (WoS) and Scopus databases to conduct our search because they are well-established and widely recognised databases, offering extensive coverage across disciplines (i.e. business, economics and management, information systems) dating back to 1945 and 1966, respectively (Zhu and Liu, 2020; Cabeza-Ramírez et al., 2021). We performed our search using the search query [“live-stream” OR “live stream” OR “live-streaming” OR “live streaming”] AND [“marketing”] AND [“purchase”] OR [“purchase intention”]. We restricted the search to years 2012-2024. In terms of selection, we only included peer-reviewed journal papers that were written in English.
The first search yielded 1,144 (1,013 from WoS and 131 from Scopus) papers, of which 309 papers were deemed relevant. To validate the comprehensiveness of our search, we cross-referenced these papers with previous SLRs in related areas and found substantial overlap, reinforcing the robustness of our approach. We then followed Zhang and Benyoucef (2016)’s approach to cross-check and validate the relevance of these papers. The first three authors independently screened the titles, abstracts and keywords against the search criteria to identify the relevant papers for the review. The agreement rate among the authors was 98%. Discrepancies were resolved through discussion with the last author. Finally, the first author performed a full-text review of the selected studies, resulting in 93 research papers for review.
2.2 Papers’ characteristics
This study was further augmented by a quantitative frequency evaluation using key research focus and context following Fu et al. (2024). We observed an increased quantity of research on influencer marketing in general, as shown in Figure 2. Further, researchers have devoted closer attention to live-stream influencer marketing in recent years. The 93 included studies were diverse in their aims, research approaches, and evaluation designs. Figure 3 provides an overview of all included studies, specifying concise information on the specific research approach.
The graph labeled “Live-stream papers (related to our R Qs)” shows years in the horizontal axis ranging from 2010 to 2026 in increments of 2 years. The vertical axis shows percentages ranging from negative 10 percent to 40 percent in increments of 5 percent. The upward-sloping curve begins near (2012, 1 percent), passes through (2013, 1 percent), (2014, 2 percent), (2015, 2 percent), (2016, 3 percent), (2017, 2 percent), (2018, 7 percent), (2019, 11 percent), (2020, 14 percent), (2021, 28 percent), (2022, 30 percent), (2023, 34 percent), and ends near (2024, 33 percent). A dotted trend line with the equation y equals 0.0319 times x minus 64.242 and R squared equals 0.8569 slopes upward from the point (2012, negative 6 percent) and ends at the point (2024, 32). Note: All numerical data values are approximated.Live stream studies’ trends. Source: Authors’ own creation
The graph labeled “Live-stream papers (related to our R Qs)” shows years in the horizontal axis ranging from 2010 to 2026 in increments of 2 years. The vertical axis shows percentages ranging from negative 10 percent to 40 percent in increments of 5 percent. The upward-sloping curve begins near (2012, 1 percent), passes through (2013, 1 percent), (2014, 2 percent), (2015, 2 percent), (2016, 3 percent), (2017, 2 percent), (2018, 7 percent), (2019, 11 percent), (2020, 14 percent), (2021, 28 percent), (2022, 30 percent), (2023, 34 percent), and ends near (2024, 33 percent). A dotted trend line with the equation y equals 0.0319 times x minus 64.242 and R squared equals 0.8569 slopes upward from the point (2012, negative 6 percent) and ends at the point (2024, 32). Note: All numerical data values are approximated.Live stream studies’ trends. Source: Authors’ own creation
The data from the pie chart in the clockwise sense are as follows: Other: 1 percent. Qualitative: 30 percent. Quantitative: 69 percent.Common research approaches. Source: Authors’ own creation
The data from the pie chart in the clockwise sense are as follows: Other: 1 percent. Qualitative: 30 percent. Quantitative: 69 percent.Common research approaches. Source: Authors’ own creation
We further classified the selected papers according to the platforms used in live-stream influencer marketing, as shown in Figure 4. We noted that Instagram was used the most. The category “others” here encompassed platforms that were used only once across the studies, such as personal blogs, Douyu and Kuaishou. We also examined previous research’s focus areas and found that lifestyle has received much attention in the live-stream marketing spectrum, as can be seen in Figure 5. Note that the total count shown in Figures 4 and 5 exceeds 93, as some studies used more than one platform and focused on more than one area.
The bar graph has the heading “Platforms Used.” The markings on the horizontal axis from left to right are “Instagram,” “No Specific Tools,” “YouTube,” “Others,” “Facebook,” “Twitch,” “Twitter,” “Taobao,” and “Weibo.” The data from the bars on the graph is as follows: Instagram: 38 percent. No Specific Tools: 19 percent. YouTube: 12 percent. Others: 11 percent. Facebook: 10 percent. Twitch: 8 percent. Twitter: 4 percent. Taobao: 4 percent. Weibo: 3 percent.Platforms used (multiple response). Source: Authors’ own creation
The bar graph has the heading “Platforms Used.” The markings on the horizontal axis from left to right are “Instagram,” “No Specific Tools,” “YouTube,” “Others,” “Facebook,” “Twitch,” “Twitter,” “Taobao,” and “Weibo.” The data from the bars on the graph is as follows: Instagram: 38 percent. No Specific Tools: 19 percent. YouTube: 12 percent. Others: 11 percent. Facebook: 10 percent. Twitch: 8 percent. Twitter: 4 percent. Taobao: 4 percent. Weibo: 3 percent.Platforms used (multiple response). Source: Authors’ own creation
The bar graph has the heading “Focus Area.” The markings on the horizontal axis from left to right are “Lifestyle,” “Business,” “Entertainment,” “Tourism,” “Others,” “Food,” “General Ad,” and “Electronic.” The data from the bars on the graph is as follows: Lifestyle: 45 percent. Business: 15 percent. Entertainment: 11 percent. Tourism: 10 percent. Others: 8 percent. Food: 6 percent. General Ad: 5 percent. Electronic: 3 percent.Focus areas (multiple responses). Source: Authors’ own creation
The bar graph has the heading “Focus Area.” The markings on the horizontal axis from left to right are “Lifestyle,” “Business,” “Entertainment,” “Tourism,” “Others,” “Food,” “General Ad,” and “Electronic.” The data from the bars on the graph is as follows: Lifestyle: 45 percent. Business: 15 percent. Entertainment: 11 percent. Tourism: 10 percent. Others: 8 percent. Food: 6 percent. General Ad: 5 percent. Electronic: 3 percent.Focus areas (multiple responses). Source: Authors’ own creation
2.3 Research questions
We adopted the PIOC (Population, Interest, Outcome, and Context) framework, as outlined by Hasan and Bao (2021), to develop focused research questions. In the context of this SLR, the population refers to online consumers, interest covers the underlying theories, outcome includes the motivational factors, and context is live-stream marketing. The primary research questions guiding our review are:
Which theories (interest) are predominantly applied in live-stream influencer marketing (context) research?
What are the key motivational factors (outcome) for viewers (potential consumers, which are the population), and how do these influence consumer attitudes and behaviours, including purchase intention?
3. Results
3.1 Empirical models and frameworks in influencer marketing (RQ1)
Numerous influencer marketing studies have applied different theories to investigate how live-stream marketing affects consumers’ purchase decision-making processes (Wu and Yu, 2024; Li et al., 2023; Xue et al., 2025; Long et al., 2024b; Wang et al., 2020; Lim et al., 2020; Khan and Rundle-Thiele, 2019). Table 1 presents an overview of related theoretical models and frameworks.
Overview of influencer marketing theories
| Live-stream influencer marketing theories | Sources | Mediator/ Moderator | Outcome |
|---|---|---|---|
| Stimulus-organism-response (SOR) (Yang et al., 2024; Zhang et al., 2023b; Li et al., 2023; Huang, 2024; Han et al., 2024; Tang et al., 2024a, b; Lee and Wan, 2023; Tan, 2024; Du and Huang, 2023; Wu and Huang, 2023; Meng and Lin, 2023) Enhanced SOR (Mutambik, 2024) stimulus-organism-behaviour-consequences (SOBC) (Yang et al., 2024; Han et al., 2024) | Trust, social presence, interactivity, parasocial relationship, values, affordance, word-of-mouth, utilitarian, price, uncertainty reduction, platform characteristics, word-of-mouth, service quality dimensions, satisfaction, loyalty, trust, engagement | Attitudes towards influencers and products, Presence and Perceived Trust | Behavioural aspect |
| Consumer engagement theory (Han et al., 2024) | Trust, interactivity | Purchase intention | |
| Sensory marketing theory and Cognitive appraisal theory (Tang et al., 2024b) | Cognitive closure, preference | Need for cognitive closure (moderator) | Behavioural intention |
| Social cognitive theory and technology acceptance model (Kwangsawad et al., 2024) | Brand awareness, content valence of viewers' attitudes | Purchase intention | |
| Community gift-giving model (Hsieh et al., 2023) | Competitive arousal, gift design aesthetics and broadcaster’s image | Chinese impression management (mianzi) (Self-mianzi, mutual mianzi - moderator) | Impulse buying behaviour |
| Quality loyalty model (Hardiyanto et al., 2024) | Parasocial relationships, and uses and gratifications | Satisfaction and loyalty | |
| Heuristic-systematic model (Zhang et al., 2023a) | Information overload, perceived product quality and fit, and streamer influence and expertise | Purchase decision | |
| Howard-sheth model (Chen and Wu, 2024) | Expertise, trustworthiness, attraction, interactivity, and affinity, Cultural trait (social orientation and face consciousness) | Purchase of virtual gifts | |
| Source credibility and source attractiveness models (Cheng et al., 2024) | Credibility and attractiveness attributes, trustworthiness, government credibility, physical attractiveness, interaction friendliness | Perceived security and enjoyment (mediator) | Purchase intention and local brand awareness |
| Triple-path model (Hong et al., 2024) | Social presence | Behavioural and emotional engagement (mediator) | Purchase behaviour |
| Parasocial interaction theory (Long et al., 2024a, b; Shen et al., 2022) | Parasocial interaction, self-accountability, viewers’ self-congruence and value congruence | Parasocial interaction (moderator), Emotional interaction (mediate) | (Green) purchase intention |
| Social presence theory and information overload theory (Cao et al., 2022) | Information overload, social presence, risk, interactivity, immediate interaction anxiety, verbal intimacy, and virtual physical intimacy | Information overload (moderator) | Purchase expensive virtual gifts |
| Social power theory (Wang et al., 2020) | Expert power | Trust, Commitment Work performance | Psychological or behavioural aspects |
| Social cognitive theory (Lim et al., 2020)* | Personal factor, environmental factor, behavioural factor | Behavioural aspects | |
| Social identity theory (Hsu, 2022; Zhou et al., 2022) | Opinion leader participation, number of relationship, links | Tipping Frequency | |
| Trust transfer theory (Hsu, 2022) | Social identification, motivation, community instructiveness | Psychological contract | Repurchase intention |
| Attitude intention behaviour theory (Khan and Rundle-Thiele, 2019) | Reasonable action | Intention | Behavioural aspects |
| Conceptual model applying scale development process (Liu et al., 2022) | Socialising, media engagement, product examination | Behaviour intention User satisfaction | |
| Two-step flow of communication theory (Katz, 1957) | Expert, peer | Behavioural aspects | |
| Information integration theory (Nascimento et al., 2020) | Information online | Observing others (influencers) and forming attitudes | Intention and behavioural aspects |
| Theory of planned behaviour (Chetioui et al., 2020) | Attitude, subjective norms, behavioural control | Intention | Behavioural aspects |
| Technology acceptance theory (Hu et al., 2019) | Consumer perceived usefulness, consumer perceived ease of use | Trust | Intention and behavioural aspects |
Source(s): Authors’ own creation
Our SLR reveals that theories—such as the Stimulus-Organism-Response (SOR), Extended SOR, SOR-Consequences, Consumer Engagement Theory, Sensory Marketing Theory, Parasocial Interaction Theory, Social Cognitive Theory (SCT), Community Gift-Giving Model, and Quality Loyalty Model—have been applied in the live-stream influencer marketing context, as summarised in Table 1. Among these, SOR remains the dominant framework to explain consumers’ behaviours, as it effectively captures the influence of different environmental stimuli (Mutambik, 2024; Yang et al., 2024a; Han et al., 2024; Xu et al., 2020; Chen et al., 2022). Researchers have identified that these stimuli in live-stream influencer marketing can trigger consumers’ emotional and cognitive processes and result in certain behaviours (Liang et al., 2024; Huo et al., 2023). Lim et al. (2020) used SCT to predict diverse human behaviours, such as health and learning behaviours, within the live-stream context. Kwangsawad et al. (2024) proposed an SCT-based research model to investigate ongoing purchase intentions for community enterprises, highlighting its role in driving consumer engagement and sales. However, SCT states that human behaviour is shaped by personal (perceived values and outcome expectations), environmental (interactions and others’ behaviours) and behavioural (prior behaviours) factors. Thereby, SCT posits that human behaviour is driven by both viewers’ internal cognition and external influences.
A recent trend in live-stream marketing research is the development of conceptual models that enhance or modify the conventional information system theory. Following Singh and Sahu’s (2020) classification approach, this SLR identified key studies (Yang et al., 2024a; Han et al., 2024; Mutambik, 2024) that propose conceptual and hybrid frameworks to examine purchasing intentions. These classifications, illustrated in Figure 6, highlight motivation, attitude, and behaviour (MAB) as the main fields within customer decision-making in the context of social media influencer marketing. While the PIOC framework is fundamentally research-driven, MAB focuses on behaviour, explaining how internal and external motivations influence attitudes and behaviours in marketing. Unlike SOR, which emphasises external stimuli, MAB integrates motivation as a key impact on attitudes and behaviours, offering a more comprehensive perspective on cognitive and emotional influences.
The flowchart is divided into three main sections arranged horizontally from left to right, labeled “Motivation,” “Attitude,” and “Behavior.” Within the “Motivation” section on the left, three vertically stacked ovals appear. The top oval is labeled “Consumers’ Motivations,” and lists the following elements: 1. Utilitarian, 2. Hedonic, 3. Symbolic Value, 4. Parasocial Interaction, 5. Trustworthiness, 6. Self-Congruence, 7. Word-of-Mouth, 8. Uncertainty Reduction, and 9. Price. Below it, the second oval is labeled “Influencers’ Motivation,” and contains: 1. Verbal Intimacy, 2. Interactivity, 3. Interaction Friendliness, and 4. Gift Design Aesthetics. At the bottom, a third oval includes “Virtual Physical Intimacy” and “Cultural Trait.” All three ovals are enclosed in a large dotted rectangle. An arrow from the Motivation section points to a rectangle labeled “Mediators,” which lists: 1. Social Presence, 2. Cognitive Trust, 3. Affective Trust, 4. Attitudes Towards Influencers and Products, 5. Emotional Interaction, and 6. Perceived Security and Enjoyment. This “Mediators” box connects to the “Attitude” section in the center. Inside, two vertically aligned ovals are present. The top oval is labeled “Source Related Factors,” and includes: 1. Source Credibility (Sources Attractiveness, Expertise, Trustworthiness), 2. Information Quality, 3. Para Social Interactions, 4. Viewer’s Content Valence, 5. Attraction, and 6. Other Aspects of Source Related Factors (that is, Government Credibility). The bottom oval is labeled “Congruence Related Factors,” and includes: 1. Self-Influencer Congruence (Transfer Model, Wishfulness Identification, Face Consciousness), 2. Product-Influencer Congruence (Match-up Hypotheses), and 3. Reduction of Information Overload. Both ovals are enclosed in a large dotted rectangle. Individual arrows from both the ovals in the attitude section lead to the “Behavior,” section, which contains a list of behavioral outcomes: “Purchase Intention,” “Actual Purchase,” “Impulse Buying Behavior,” “Consumer Satisfaction,” “Consumer Loyalty,” “Brand Awareness,” “Green Purchase Behavior,” and “Repurchase Intention.” Above the Behavior section, a dotted oval labeled “Moderators” includes the following items: 1. Need for Cognitive Closure, 2. Impression Management (Self or Mutual), 3. Para-Social Interaction, 4. Information Overload, and 5. Psychological Contract. A downward arrow from this oval points to the arrow leading to the “Behavior section,” from the ovals in the attitude section.A framework of consumer decision-making in live-stream influencer marketing. Source: Authors’ own creation
The flowchart is divided into three main sections arranged horizontally from left to right, labeled “Motivation,” “Attitude,” and “Behavior.” Within the “Motivation” section on the left, three vertically stacked ovals appear. The top oval is labeled “Consumers’ Motivations,” and lists the following elements: 1. Utilitarian, 2. Hedonic, 3. Symbolic Value, 4. Parasocial Interaction, 5. Trustworthiness, 6. Self-Congruence, 7. Word-of-Mouth, 8. Uncertainty Reduction, and 9. Price. Below it, the second oval is labeled “Influencers’ Motivation,” and contains: 1. Verbal Intimacy, 2. Interactivity, 3. Interaction Friendliness, and 4. Gift Design Aesthetics. At the bottom, a third oval includes “Virtual Physical Intimacy” and “Cultural Trait.” All three ovals are enclosed in a large dotted rectangle. An arrow from the Motivation section points to a rectangle labeled “Mediators,” which lists: 1. Social Presence, 2. Cognitive Trust, 3. Affective Trust, 4. Attitudes Towards Influencers and Products, 5. Emotional Interaction, and 6. Perceived Security and Enjoyment. This “Mediators” box connects to the “Attitude” section in the center. Inside, two vertically aligned ovals are present. The top oval is labeled “Source Related Factors,” and includes: 1. Source Credibility (Sources Attractiveness, Expertise, Trustworthiness), 2. Information Quality, 3. Para Social Interactions, 4. Viewer’s Content Valence, 5. Attraction, and 6. Other Aspects of Source Related Factors (that is, Government Credibility). The bottom oval is labeled “Congruence Related Factors,” and includes: 1. Self-Influencer Congruence (Transfer Model, Wishfulness Identification, Face Consciousness), 2. Product-Influencer Congruence (Match-up Hypotheses), and 3. Reduction of Information Overload. Both ovals are enclosed in a large dotted rectangle. Individual arrows from both the ovals in the attitude section lead to the “Behavior,” section, which contains a list of behavioral outcomes: “Purchase Intention,” “Actual Purchase,” “Impulse Buying Behavior,” “Consumer Satisfaction,” “Consumer Loyalty,” “Brand Awareness,” “Green Purchase Behavior,” and “Repurchase Intention.” Above the Behavior section, a dotted oval labeled “Moderators” includes the following items: 1. Need for Cognitive Closure, 2. Impression Management (Self or Mutual), 3. Para-Social Interaction, 4. Information Overload, and 5. Psychological Contract. A downward arrow from this oval points to the arrow leading to the “Behavior section,” from the ovals in the attitude section.A framework of consumer decision-making in live-stream influencer marketing. Source: Authors’ own creation
3.2 Motivation factors, consumer attitudes and behaviours (RQ2)
The theoretical models and frameworks reviewed above indicate that consumers are first motivated by live-streaming, which shapes their attitudes towards live-stream marketing and products, ultimately influencing their purchasing behaviour. While we acknowledge that motivation is not the sole factor influencing consumer attitudes and intentions, our study focuses on examining the specific impacts of motivation in live-stream marketing contexts. We would also like to point out that motivation here denotes the set of intrinsic and extrinsic forces that (a) prompt consumers to engage with and purchase through live-stream content and (b) drive influencers to create, curate, and disseminate that content. Together, these forces shape message quality and authenticity, influence consumer attitudes, and activate purchase behaviour.
3.2.1 Motivation
3.2.1.1 Consumers’ motivation
Xu and Ye (2020) identified three primary motivations for live-stream viewers: entertainment, information seeking and socialisation. Cabeza-Ramírez et al. (2022), on the other hand, classified utilitarian, hedonic and symbolic values as motivators for viewers to watch live streams.
Utilitarian value, the utility that consumers wish to gain, is the strongest drive of consumers’ purchase intention and behaviour (Wongkitrungrueng and Assarut, 2020; Bu et al., 2022; Zhang and Wang, 2025). Hedonic value reflects the degree of enjoyment consumers perceive (Chen et al., 2024c). The effect of presenting entertainment in e-commerce can lead to consumers’ willingness to view online stores (Zhang and Zhang, 2024). Symbolic value fosters social integration (Chen et al., 2022; Koo et al., 2008; Tan, 2024), self-fulfilment, and relationship building, influencing purchase intention (Wongkitrungrueng and Assarut, 2020; Liu et al., 2022; Peng et al., 2024; Lawrence and Meivitawanli, 2023).
Several factors also influence consumers’ motivations. Kwangsawad et al. (2024), Zhang et al. (2023a, b), Arnocky et al. (2016) and Chen et al. (2022) indicated that benign envy strengthens consumers’ attitudes towards desired products and motivates the use of social media e-commerce. In addition, higher materialism increases satisfaction, leading to positive attitudes towards products (Lou and Kim, 2019).
A key advantage of live-stream marketing is that it allows consumers to see the seller’s actual appearance in real-time (Zhang, 2023), enabling consumers’ authentic experience (Wongkitrungrueng and Assarut, 2020; Zhong et al., 2022; Gong et al., 2022). Further, various live-stream contents enhance viewer enjoyment while satisfying information needs (Xu and Ye, 2020; Liu et al., 2022; Lin and Lee, 2024). Live-stream influencer marketing’s interactive, dynamic nature has successfully combined utilitarian, hedonic and symbolic values.
3.2.1.2 Influencers’ motivation
Understanding influencer motivation is essential, as it directly impacts the quality, authenticity, and type of information shared in live-stream marketing, influencing consumer perceptions, attitudes, and subsequent behaviours (Quelhas-Brito et al., 2020; Rungruangjit, 2022; Lee and Wan, 2023; Lawrence and Meivitawanli, 2023). Self-Determination Theory (SDT) explains these motivations, distinguishing extrinsic motivations such as financial rewards from intrinsic motivations of hedonic experience seeking (Chen et al., 2023; Yu and Zheng, 2022).
In addition, SDT proposes three conditions of psychological needs—competency, relatedness and autonomy—that significantly affect influencers’ interactions with their audiences. Competency reflects the desire to be seen as knowledgeable, relatedness refers to connections with viewers, and autonomy allows influencers to maintain creative control over their content (Ryan and Ryan, 2019). By exploring these motivational factors, SDT provides deeper insights into how influencer-driven contents can effectively shape consumer motivations, attitudes, and ultimately their purchase decisions in live-stream marketing contexts.
3.2.2 Consumer attitude
The live-stream influencer marketing literature on consumer attitude and endorsement effect can be generally classified into two categories: source-related and congruence-related studies. While multiple factors influence consumer attitudes, concentrating on source-related perception and congruence perception allows for a deeper exploration of their specific roles and dynamics in influencer marketing.
3.2.2.1 Source-related perception
The source credibility model is widely used to investigate live-stream influencer effects on consumer decision-making (Fink et al., 2020; Zhou et al., 2022; Dokumaci, 2024). The model suggests that influencers’ credibility affects consumers’ acceptance of an endorsed message (Shan et al., 2020; Rungruangjit, 2022; Ji, 2024). Consumers tend to believe that live-stream influencers are more credible than traditional celebrities (Gong et al., 2022). However, research remains divided on source credibility’s influence on consumer attitudes and purchase intentions (Lee and Kim, 2020).
Table 2 presents an overview of related source credibility studies. Several studies (e.g. Sokolova and Kefi (2020), Reinikainen et al. (2020), Rungruangjit (2022), Lawrence and Meivitawanli (2023), Yen et al. (2024) and Jiang et al. (2022)) have adopted Ohanian (1990)’s source credibility model, which includes source attractiveness, expertise and trustworthiness, to explain its influence on consumers’ behaviour.
Source credibility model
| Studies | Dimensions | Mediator | Outcomes |
|---|---|---|---|
| Rungruangjit (2022), Lee and Kim (2020), Pick (2021), Jin and Muqaddam (2019), Jin et al. (2019), Thomas and Johnson (2019), Sokolova and Kefi (2020), Reinikainen et al. (2020), Breves et al. (2019), Lawrence and Meivitawanli (2023), Yen et al. (2024), Jiang et al. (2022) | Source attractiveness Source expertise Source trustworthiness | Consumers’ attitude | Consumers’ intention and behaviour |
| Lou and Kim (2019), Park and Lin (2020), Shan et al. (2020), Lou and Kim (2019), Shan et al. (2020) | Source expertise Source trustworthiness | Consumers’ attitude | Consumers’ intention and behaviour |
Source(s): Authors’ own creation
Source attractiveness refers to familiarity and similarity of the source, affecting consumers’ attitudes, advertising effectiveness, and perceived influencer popularity (Park and Lin, 2020; Taillon et al., 2020; Shan et al., 2020; Rungruangjit, 2022). Studies showed that the attractiveness of live-stream influencers can increase the endorsement effectiveness (Niu and Ma, 2025; Park and Lin, 2020) and positively affect product attitudes and purchase intentions (Pick, 2021; Guo et al., 2022). Source expertise is another critical factor affecting credibility, which refers to the extent of valid assertions of a source perceived by consumers (Shan et al., 2020; Sokolova and Kefi, 2020).
Source trustworthiness refers to consumers’ perceptions of live-stream influencer integrity and honesty (Park and Lin, 2020). Research indicates that trustworthiness enhances influencer effectiveness, especially for consumers’ purchase intentions (Alnoor et al., 2024). When brand endorsements do not require expertise, trustworthiness becomes the determinant of an influencer’s impact (Ji, 2024).
3.2.2.2 Congruence
Congruence-related studies in influencer marketing focus on two aspects—the congruence between products and live-stream influencers, and the congruence between consumers’ ideal self-images and live-stream influencer images (Shan et al., 2020; Lawrence and Meivitawanli, 2023).
For the former, researchers have found that the degree of the match-up between live-stream influencers and products will affect consumers’ attitudes and purchase intentions (Park and Lin, 2020; Monge-Benito et al., 2020; Breves et al., 2019; Lu and Chen, 2021; Hsu, 2022). A message’s credibility and persuasiveness depend on the level of match-up and consistency between the live-stream influencer and the attributes of the brand or product (Monge-Benito et al., 2020; Rungruangjit, 2022). Park and Lin (2020), Chen et al. (2024b) and Jin and Muqaddam (2019) further identified that live-stream influencers can strengthen the endorsement effect when they advertise products in a more natural setting.
For the latter, acquiring the desired attributes and symbolic meaning from live-stream influencers, consumers start to construct their self-images, a process known as wishful identification (Shan et al., 2020). Researchers have shown that followers with stronger wishful identification develop greater loyalty and long-term relationships with live-stream influencers. Schouten et al. (2020) stated that wishful identification derives from both actual and perceived similarity, with live-stream influencers fostering stronger wishful identification than traditional celebrities (Wahab et al., 2022).
3.2.3 Consumer behaviour
3.2.3.1 Consumer engagement (CE)
Consumer behaviour in influencer marketing is defined as consumers’ cognitive, emotional and behavioural participation in some engagement activities (Hughes et al., 2019; Liu et al., 2022). A higher degree of CE strengthens the perceptions of live-stream influencers and subsequently triggers actual behaviour (Villamediana-Pedrosa et al., 2019; Zheng et al., 2022; Niu and Ma, 2025). Researchers have categorised CE into two dimensions: psychological aspects, such as satisfaction and commitment to events, and behavioural aspects, including consumers’ value-adding behaviours (Bergel et al., 2019; Ding et al., 2024; Chen and Yang, 2023).
Giakoumaki and Krepapa (2020) suggested that CE focuses on the consumer-brand relationship and explains consumers’ interactive experiences with the “engagement objects”. In social-commerce, different platforms provide various engagement opportunities with specific brands or live-stream influencers, who function as human brands. Consumers engage with live-stream influencers to seek entertainment or hedonic experiences, social connection and self-expression (Zhang and Zhang, 2024; Chen et al., 2024a).
Bergel et al. (2019) classified consumer engagement behaviour as transactional and non-transactional behaviour. Transactional behaviour captures consumer purchasing behaviour that can directly add to the brand’s value. Non-transactional behaviour includes customer referral behaviour (CRB) (or eWOM behaviour (Ko and Ho, 2024)), customer knowledge behaviour (CKB) and consumer influencer behaviour (CIB). Each of these generates different values; for example, CRB generates consumer referral value, CKB generates consumers’ knowledge value, and CIB the influence value (Bergel et al., 2019; Khare et al., 2022).
3.2.3.2 Consumer purchase intention
Empirical research confirmed that social media significantly affects purchase intentions (Song et al., 2024; Li and Gao, 2024; Zhang et al., 2023a; Yin et al., 2023; Villamediana-Pedrosa et al., 2019; Wahab et al., 2022; Shan et al., 2020), which is considered as the critical outcome to determine the effect of social media endorsement and actual purchasing behaviour (Kay et al., 2020; Chetioui et al., 2020; Taillon et al., 2020; Wahab et al., 2022). Ho et al. (2022), Yang et al. (2024b), and Cabeza-Ramírez et al. (2022) found that consumer attitudes towards live-stream influencer marketing directly predict purchase intentions. Fink et al. (2020) further indicated that live-stream influencer endorsements influence behavioural intentions.
4. Discussion
This paper advances the live-stream influencer marketing literature by comparing and synthesising the effects of social media and live-stream influencer endorsements on consumers’ decision-making processes. It examines key motivators, including consumer motivations, attitude shifts, and purchase intentions, from the perspectives of influencers, consumers, and content. Additionally, it explores consumer values perceived in influencer marketing and their impact on behaviour. By providing a comprehensive literature review, this study identifies key factors and theoretical frameworks influencing consumer perceptions and purchase intentions.
From the consumer perspective, this SLR highlights that the endorsement process begins with consumers' motivation to engage in live-stream influencer marketing, driven by utilitarian, hedonic, and symbolic values. Engagement leads consumers to form attitudes toward live-stream influencers and brands, influenced by similarity, likeability, subjective norms, and self-image congruence, and relationships with live-stream influencer (PSI and PSR). We then examined two key outcomes: CE and purchase intentions. CE encompasses psychological aspects, like satisfaction and commitment (Hughes et al., 2019; Li and Gao, 2024), and behavioural aspects, such as actions that support brand activities and influence purchase intentions, the primary measure of influencer marketing effectiveness (Chetioui et al., 2020; Bergel et al., 2019; Bu et al., 2022; Zhang et al., 2024b; Lee and Kim, 2020).
From the influencer’s perspective, this SLR summarises the motivations driving social media influencers to share contents, classified as intrinsic (hedonic-driven) and extrinsic (reward-seeking) motivations (Chen et al., 2023; Quelhas-Brito et al., 2020). We also examined characteristics of live-stream influencer from source-related and congruence-related studies that influence consumer attitudes and purchase intentions. Additionally, congruence between the advertised product and live-stream influencer fosters favourable attitudes toward both the products and live-stream influencer (Yang et al., 2024b; Ho et al., 2022), ultimately triggering purchase intentions.
4.1 Theoretical and practical implications
This SLR offers several significant theoretical and practical implications. Firstly, it consolidates social media live-stream influencer marketing research, an emerging area with unique characteristics, from two perspectives: influencers and consumers. Synthesising 93 studies, this SLR identifies critical research gaps, such as the limited application of live-stream-specific theories and highlights areas requiring further exploration. Secondly, this SLR consolidated key theories including the SOR framework and SCT, that underpin consumer behaviour in the context of influencer marketing. Thirdly, this SLR elucidates how motivational factors such as utilitarian, hedonic, and symbolic values influence consumer attitudes, engagement, and purchase intentions. The classification approach adopted highlights motivation, attitude, and behaviour as central domains, offering a structured lens for future research in the area of customer decision-making in live-stream contexts.
From practitioners’ perspectives, this SLR offers actionable insights for optimising live-stream marketing strategies. Marketers should prioritise live-stream influencer-consumer similarity, as this effectively trigger purchase intentions. Facilitating consumers-influencer real-time interaction in live streams strengthens emotional connections and trust.
5. Limitations and future research directions
This study has a few limitations. First, it focuses solely on key factors influencing consumer attitudes and intentions. Future research could benefit from exploring additional variables such as social norms, personal experiences, and contextual variables, as well as their interaction with motivation. Secondly, this study exclusively sourced only peer-reviewed papers from the WoS and Scopus databases, ensuring high-quality material but potentially excluding important information from other sources. Future research could enhance the scope by reviewing papers published in leading journals across multiple databases.
Additionally, existing influencer marketing studies often overlook the impact of cultural differences. Consumers in different regions may seek distinct values from social media commerce; for instance, East Asian consumers often prioritise Confucian values (Li et al., 2023), while Western consumers may focus more on hedonic and utilitarian values (Wongkitrungrueng and Assarut, 2020). Future research should conduct more comprehensive cross-cultural comparisons to better understand consumer behaviour across diverse cultural contexts.
Another limitation of previous research is the lack of segmentation in consumer behaviours, particularly purchase intentions. Some products and services, like unique restaurant dishes or tourist attractions, require in-person experiences and cannot be purchased online. These high-experience products necessitate physical store visits or virtual reality (VR) content creation, making it important to distinguish between online and offline purchase intentions.
Finally, this study recommends emerging avenues for future research. One promising direction is improving the consumer experience in live-stream marketing. Investigating novel methodologies for employing AI-assisted (i.e. VR) live-streaming could enhance the personalisation and engagement of the user experience. This approach may facilitate an investigation into the role of augmented reality and VR in improving live-stream interactions, thereby enabling users to visualise products more effectively and simulate offline experiences.
Furthermore, it is likely to conduct research to assess the efficacy of limited-time offers and flash sales in eliciting FOMO (fear of missing out) and shaping purchase intentions. Moreover, it is recommended to examine the ways in which live-stream influencers can adeptly promote eco-friendly products and convey sustainability values. It is of paramount importance that we conduct a thorough investigation into the long-term revenue streams associated with the streaming economy. Future research should investigate the potential of live-stream marketing to establish enduring revenue streams for enterprises by means of continuous engagement, subscription-based sales models, and repurchase intentions.
In addition, illogical and unintentional purchasing behaviour can occasionally hinder consumers from making purchases during live-streaming marketing events. Future research should examine the influence of emotions and impulsive decision-making in driving unplanned purchases in live-stream settings. In this instance, enterprises could implement a gamification strategy and subsequently evaluate the impact of gamification and reward systems on enhancing impulsive buying decisions.
This study further advocates for future research into the application of AI to improve real-time product recommendations and chatbot interactions in the context of live-stream events. Another exciting avenue for future research is the exploration of gift-giving marketing within live-stream contexts. Research may be undertaken to explore the psychological triggers that influence gift-giving motives during live-stream events, encompassing cultural, emotional, and social factors. Finally, we put forth the idea of formulating green marketing strategies specifically tailored for live-streams, wherein we examine the various storytelling techniques that live-stream influencers can employ to exhibit the sustainability and eco-friendliness of products effectively. Collectively, these study directions tackle the expanding possibilities of live-stream influencer marketing and complement emerging digital marketing technology, sustainability objectives, and consumer behaviour patterns.
6. Concluding remarks
This systematic literature provides important insights into the emerging field of live-stream influencers on consumer behaviour and attitudes. By examining the perspectives of both influencers and consumers, this review highlights key findings from previous research related to the motivational factors driving consumer engagement and the theoretical frameworks underpinning influencer marketing strategies. Additionally, it calls for the development of live-stream specific theoretical frameworks to address the unique dynamics of real-time interaction and emotional engagement. These insights offer valuable directions for researchers and actionable recommendations for practitioners seeking to optimise live-stream influencer marketing strategies. While restricting the review to the Web of Science database presents a limitation, the identified research gaps underscore the need for further exploration in this dynamic field. Notably, this review is among the first to examine the effects of live-stream influencer marketing on consumer purchase intentions, offering a foundational perspective for future studies to expand and deepen our understanding of this evolving area.
The authors would like to acknowledge funding support from the Business and Industry and University of Newcastle International Postgraduate Research Scholarship. The authors would also like to acknowledge funding support from industry partners Roboworks Pty Ltd and Max on Wines Pty Ltd.
Declaration: The authors declare no conflict of interest.
