Introduction

Amid the rapid development of the internet, China has entered an era where all people are online. The Statistical Report on Internet Development in China has indicated that as of December 2022, the number of Internet users in China reached 1.067 billion (CNNIC, 2023). Technological innovation and development in the field of digital information have generated a series of short video application platforms. In terms of the fast-growing network entertainment lifestyle, short videos have become an emerging popular culture and social communication language. Similar to Instagram, Snapchat Story, Houseparty, and Micodeng are owned by Twitter and Facebook. TikTok, Kwai, and Volcano videos are customized based on the usage habits of Chinese users and have become the main components of the domestic short video ecosystem. As of December 2022, the number of short video users reached 1.012 billion (CNNIC, 2023). Short video marketing meets the entertainment needs of people in regard to their busy lives because of its low threshold and fast entertainment (Louisa and Yang, 2023). Short video content consumption has also successfully broken through barriers and entered the public’s field of vision.

Different from static pictures or traditional videos, short videos have a low entry barrier, rich information, and fast transmission speed, which can fully arouse consumers’ virtual senses (Xiao et al., 2020). Bhardwaj and Sharma (2020) indicated that short videos can activate vision and hearing simultaneously, thereby fully activating the human brain, and are easier for audiences to remember. Sun (2020) showed that short video marketing can enhance the authenticity of video content and make audiences feel more interactive. Sedej (2019) also found that short video marketing has high information density, which is conducive to awakening the emotional needs of an audience and creating conditions for direct communication between enterprises and audience members. On the psychological level of users, Cheng and Li (2023) found that short videos like TikTok news with negative sentiment and more second-person perspectives were associated with higher viewer engagement. In the research regarding short video marketing’s impact on consumers’ purchase intention, the content characteristics of the short video itself and the impact of the short video marketing platform on consumers’ purchase intention have mainly been studied (Boman and Raijonkari, 2017; Alexandra and Scarlat, 2017).

Such studies have extensively explored the impact of various factors on purchase intention in social commerce, such as brand equity (Huang et al., 2022), social learning (Chen et al., 2017), digital influencers (Wang et al., 2021), and video creation capabilities (Cuesta-Valiño et al., 2022). However, little attention has been given to the role of persona perception in short video marketing. Persona perceptions refer to consumers’ perceptions and evaluations of the personalities, characteristics, and attributes displayed by individuals or characters in short videos (Salminen et al., 2020). Understanding how such perception shapes consumers’ purchase intention is crucial for marketers to optimize their short video marketing strategies (Gan et al., 2023). Therefore, further research is needed to reveal the direct and indirect effects of persona perception on consumers’ purchase intention, potentially, by considering factors such as authenticity, relatability, and likeability (Salminen et al., 2020). Such research could also draw upon theories from consumer psychology, the notion of social influence, and media effects to provide a comprehensive understanding of the underlying mechanisms.

In the era of big data in the 21st century, short videos have become an indispensable tool for daily entertainment and corporate marketing. Although short video marketing has many advantages over traditional marketing methods, in actual marketing, the understanding of the effectiveness of such marketing promotion is also uneven. Hence, motivated by this academic and practical significance, this study explores how users’ behaviors and individual characteristics affect their purchase intention when they have value resonance with streamer perception. Specifically, we seek to answer the following research questions:

  1. 1.

    Do consumers generate purchase intention while watching short videos?

  2. 2.

    What persona perception factors create shared value with the focal platform?

  3. 3.

    What factors affect shared value and consumers’ willingness to purchase on short video platforms?

This study takes short video platform users as the research object; through field survey data, it uses structural equation modeling (SEM) to process and empirically analyze the collected data and then explores the relevant factors that affect shared value and consumer purchase intention in short video marketing. It therefore enriches the theory on short video marketing and consumer behavior and provides salient references for relevant companies involved in short video marketing.

The introduction is followed by a literature review and hypothesis development. Then comes the research framework, followed by the data analysis and results. The fifth section is discussion and implications, as well as thoughts on future research. The research design is shown in Fig. 1.

Fig. 1: Research background.
figure 1

Research design.

Full size image

Literature review and hypothesis development

The stimulus‒organism‒response framework

This study adopted the S-O-R paradigm in framework construction for three reasons. Firstly, although the S-O-R paradigm was initially used in the field of environmental psychology, marketing research has also shown that consumers’ consumption behavior is influenced by environmental stimuli (Eroglu et al., 2003; Mazaheri et al., 2011). Chinese people are dominated by collectivism in a multicultural social environment and are particularly sensitive to various environmental factors (Shobeiri et al., 2018). Therefore, the S-O-R paradigm can be used to study the response of Chinese consumers to short video stimuli. Secondly, under the S-O-R paradigm, when consumers receive multi-sensory input from the external environment, their internal state also affects their behavior. Some studies use the S-O-R paradigm to explain consumer behavior under various external environmental stimuli, resulting in certain predictions (Russell and Mehrabian, 1974; Wu et al., 2019; Ebru et al., 2023). Therefore, the S-O-R model can be used to explain the relationship between persona perception, shared value creation, and purchase intention on short video platforms. Thirdly, with the continuous expansion and deepening of short video platforms in China, studying which persona perception can capture consumers’ true intentions, meet their internal needs, and further increase their willingness to purchase goods through short video platforms has high research and practical significance (Chai et al., 2017; Li et al., 2019; Eroglu et al., 2003; Imene and Lolwah, 2023). Therefore, this study is suitable for applying the S-O-R paradigm to examine the relationship between persona perception and purchase intention on short video platforms. Finally, the S-O-R paradigm is an effective and direct approach adopted in this study to examine the impact of persona perception (S) through shared value creation (O) on consumers’ purchase intention (R) on short video platforms.

Hypothesis development

Porter and Kramer defined shared value as follows: The “policies and operating practices that enhance the competitiveness of a company while simultaneously advancing the economic and social conditions in the communities in which it operates” (Porter and Kramer, 2011). According to this definition, the interaction between short video influencers and viewers through formal channels constitutes shared value creation. Studies have shown that trust can promote interactions between influencers and viewers in e-commerce (Chang and Chen, 2008). Wongkitrungrueng and Assarut (2020) have also confirmed that customers’ trust in a seller is positively correlated with their engagement. In addition, credibility plays a crucial role in viewers’ positive attitudes toward influencers (Belanche et al., 2021). Therefore, short video influencers strengthen their credibility to gain viewers’ trust, which is conducive to promoting consumers’ participation in interaction and shared value creation. In addition, credibility has an impact on consumers’ purchase intention. Weismueller et al. (2020) found that source credibility significantly enhances consumers’ purchase intention. In a similar manner, decision use and active use intention concerning gamified e-commerce websites influence viewers’ buyback intention (Aparicio et al., 2021). Hence, short video platforms enhance their credibility through real influencers and work to ensure viewers have a positive attitude toward using them, which increases product sales and economic contributions. When viewers can create jobs for themselves by continuing to use short video platforms to make videos, they also make great contributions to society. Based on this, the following hypotheses are proposed:

H1: Credibility has a significant positive impact on shared value creation.

H2: Willingness to use has a significant positive impact on shared value creation.

Shared value creation requires platforms to convey appropriate information to viewers through short videos. Regarding the important role of information, myriad academic studies have been carried out. For example, Hartono et al. (2010) illustrated that the quality of shared information affects the performance of an operational supply chain and then improves overall firm performance. Kim et al. (2021) also confirmed that qualified information and services have a positive impact on customer satisfaction and ultimately enhance customer loyalty. This improvement in firm performance and customer loyalty can be regarded as an embodiment of shared value creation; thus, shared value creation is also positively affected by the quality of information provided by a platform. Therefore, what characteristics should information convey to have a positive impact on shared value creation? Chang (2018) proposed that high perceived information consistency is vital for the success of brands. In addition, Shamala et al. (2017) demonstrated that objectivity, completeness, reliability, and verifiability have a significant impact on the quality of information collection in information security risk management and can guide practitioners to make effective plans and reach successful decisions. Chu et al. (2018) have also used completeness, relevance, and accuracy as the criteria for judging information quality. Accordingly, based on the above research and the characteristics of short videos, we argue that the information provided by short video platforms should convey the characteristics of consistency, completeness, and clarity. Clarity is also an important factor for any platform to consider because the information provided by short video platforms should be expressed with clarity and ease, which is conducive to viewers’ understanding thereof and affects shared value creation. Based on this, the following hypotheses are proposed:

H3: Consistency has a significant positive impact on shared value creation.

H4: Completeness has a significant positive impact on shared value creation.

H5: Clarity has a significant positive impact on shared value creation.

Shared value creation is influenced not only by the platform itself but also by the emotional factors among viewers. For short video platforms, short video influencers are the most able to influence viewers’ emotions. The “similar-to-me” effect (Cotter, 2011) shows that individuals are more likely to respond to others whom they deem similar to themselves. Similarity to others helps generate empathy (Bornet and Brangier, 2016). Moreover, empathy is positively correlated with overall corporate social responsibility as well as the relevant social and environmental factors (Chourou et al., 2021). In a new relationship, customer-perceived relationship empathy also contributes to increased trust (Mangus et al., 2020). Therefore, when viewers on short video platforms find similarities with influencers, they are more likely to have a favorable impression thereof, generating empathy and contributing to shared value creation. Via empathy and similarity, viewers are better able to understand and trust influencers, better promoting stable relations between these two parties and thereby contributing to society and the economy.

Viewers’ affection for short video influencers is equally important. Akarsu et al. (2020) have demonstrated that the likability of Airbnb affects the willingness of users to visit it again and recommend it. Teng and Tsai (2020) also showed that tour leader likability at once directly affects viewers’ value cocreation behavior and indirectly affects it through the intermediary impact of tour leader attachment. Influencers with more likability on short video platforms can make viewers recognize similar aspects of themselves, reducing the sense of distance between them, which plays a positive role in shared value creation. Chen et al. (2020) have indicated the importance of research on shared value creation from a public perspective, as this contributes to the development of shared value creation from the perspectives of the public and organizations. Likability, empathy, and similarity are salient variables in the research on shared value creation from the perspective of viewers’ emotional factors. Therefore, we suggest that likability, empathy, and similarity contribute to shared value creation. Based on this, the following hypotheses are proposed:

H6: Likability has a significant positive impact on shared value creation.

H7: Empathy has a significant positive impact on shared value creation.

H8: Similarity has a significant positive impact on shared value creation.

As a new form of online shopping, shopping on short video platforms has gradually attracted the attention of consumers in recent years. To further enhance consumers’ purchase intention on short video platforms, it is particularly important to explore the antecedents thereof on short video platforms. The antecedents of continuous use intention have thus attracted academic attention. Li et al. (2018) have noted that a platform should create a warm environment that can promote emotional development between its audience and users to promote the purchase intention of the latter. Shang et al. (2017) also showed that consumer resonance on social networking sites has a significant positive impact on consumers’ purchase intention and that this phenomenon originates with sellers who release appropriate product information on social networking sites while actively interacting with consumers through appropriate channels. Fan et al. (2020) have also shown that mobile live video interactivity improves users’ identification with influencers and their experience when watching live videos, thereby promoting users’ continuous use intention. In sum, to improve users’ purchase intention, short video platforms must provide appropriate products and have sufficient formal channels for interacting with audiences and selling more products to realize economic contributions. In addition, they also must be able to stabilize their emotional relationships with audiences to create better social contributions. Such economic and social contributions embody shared value creation; thus, shared value creation generates and increases viewers’ purchase intention. Based on this, the following assumption is proposed:

H9: Shared value creation has a significant positive impact on purchase intention.

The regulatory focus has two forms: promotion focus and or prevention focus. When individuals are promotion-focused, they pay more attention to positive results. When individuals are prevention-focused, they pay more attention to the negative outcomes they want to avoid (Higgins et al., 2001). The academic community has carried out many studies based on the theory of regulatory focus. For example, Kim et al. (2012) have indicated that the influence of CSR activities on consumers’ attitudes and purchase intention depends on the type of hope inspired by CSR advertisements. While CSR advertisements that describe the negative state of the current environment and show how the contrasting improvements caused by CSR activities can trigger prevention hope, those describing a positive future image may generate promotion hope. Similarly, Zhang (2016) has found that consumers with a high degree of narcissism (promotion focus orientation) have a strong purchase intention for products presenting promotion-focused advertisements but that consumers with a low degree of narcissism (prevention focus orientation) have a strong purchase intention for products presenting prevention-focused advertisements. Thus, according to regulation focus theory and the above research conclusions, when a short video platform arouses prevention hope via shared value creation and by delivering avoidable negative information to consumers, it stimulates the purchase intention of prevention-focused consumers. When positive information is delivered to consumers and arouses promotion hope, it better stimulates the purchase intention of promotion-focused consumers. Based on this, the following hypotheses are proposed:

H10: Regulatory focus plays a moderating role between shared value creation and purchase intention.

H10a: When shared value creation reflects positive results, the purchase intention of promotion-focused consumers becomes stronger than that of prevention-focused consumers.

H10b: When shared value creation reflects negative results that must be avoided, the purchase intention of prevention-focused consumers becomes stronger than that of promotion-focused consumers.

Moreover, the mediating role of shared value creation may be moderated by social presence. When Short et al. (1976) first proposed social presence, they suggested that social presence is the degree to which a person is regarded as a “real person” and his or her perceived degree of connection with others in communication through media. The sense of presence plays an important role in communication and interaction. Similarly, social presence has an important impact on the interaction and trust between online shopping platforms and users. Due to their lack of consumer trust, many newly opened online stores end up failing. Further exploring the influence of social presence, Ogonowski et al. (2014) have found that consumers are more likely to trust a website if it offers the impression of a high level of social presence. In addition, when verifying the trust scale with the four dimensions of integrity, predictability, ability, and benevolence in the electronic product environment, Gefen and Straub (2004) also showed that social presence has a certain influence on these trust dimensions; specifically, benevolence has the strongest influence and affects online purchase intention. Ye et al. (2019, 2020) also concluded that social presence can indirectly improve customers’ trust and purchase intention through intermediary channels such as utilitarianism and hedonism. Hence, the greater the presence of a short video platform is, the greater its users’ trust due to the platform’s human touch and social sense, which stimulate viewers’ purchase intention. The lesser the presence of a platform is, the greater its users’ alienation, reducing their trust. Therefore, the quality of the interaction between the platform and the user is affected by the degree of social presence of the former, which affects the purchase intention of the latter. Hence, social presence modulates the impact of shared value creation on viewers’ purchase intention on a platform. Based on this, the following hypotheses are proposed:

H11: Social presence plays a moderating role between shared value creation and purchase intention.

H11a: The greater the sense of social presence is, the stronger the influence of shared value creation on viewers’ purchase intention.

H11b: The lesser the sense of social presence is, the weaker the impact of shared value creation on viewers’ purchase intention.

To test the validity of the research model, it is also necessary to add control variables that have an impact on purchase intention. Chen et al. (2021) have noted that personal factors (such as age, gender, income, education, impulsivity, and emotional susceptibility) affect shoppers’ impulsive purchase decisions to varying degrees due to the different types of shopping partners. In live streaming, for example, entertaining content is more likely to cause men to buy virtual gifts (Li et al., 2018). Therefore, the effects of age, gender, and income should be controlled for in the study. The content quality of the products provided by a brand on social media and the interaction between users and the company increase users’ enjoyment motivation and strengthen users’ brand awareness, especially among millennials. As a company provides increasingly high-quality web content on social media, the interaction between consumers and this company and its brand is enhanced, which attracts more consumers and establishes consumer brand awareness, thus promoting consumers’ willingness to buy offline (Dabbous and Barakat, 2020). Hence, consumers with higher brand awareness of products have a stronger purchase intention when shopping on short video platforms. Therefore, brand awareness should also be included in the model. Regarding the authors of each variable, the relevant paper information and hypotheses are shown in Table 1.

Table 1 Paper information on variables and research hypotheses.
Full size table

Research framework

This study proposes 11 hypotheses and constructs a theoretical model adopted to short video platform users, as shown in Fig. 2.

Fig. 2: Literature research.
figure 2

Research model.

Full size image

Instrument

To ensure content validity, the measurement items mainly refer to mature scales from previous research. Credibility was measured using four items adapted from Weismueller et al. (2020). Consistency was measured using four items adapted from Shamala et al. (2017). Completeness was measured using four items adapted from Kim et al. (2021). Clarity was measured using three items adapted from Chu et al. (2018). Likability was measured using four items adapted from Bornet and Brangier (2016). Empathy was measured using three items adapted from Mangus et al. (2020). Similarity was measured using four items adapted from Xiao et al. (2020). Willingness to use was measured using four items adapted from Aparicio et al. (2021). Shared value creation was measured using eight items adapted from Ham et al. (2020). Purchase intention on short video platforms was measured using three items adapted from Kim et al. (2012).

All measurement items used in this study were pretested with a sample of 20 volunteers, all of whom were short video users, similar to the target sample to be investigated. Conduct personal interviews to correct potential deficiencies and anticipate any doubts and issues that may arise during the information collection process. After the pretest, we removed the problematic indicators, and the questionnaire with 41 items was finally formulated. The final survey questionnaire is presented in Table 2. Participants involved in this pretest were not included in the main sample.

Table 2 Questionnaire items.
Full size table

Since the original text was in English, we conducted a back translation to ensure the translation validity. Firstly, a native Chinese researcher translated the source project from English to Chinese. Next, another researcher independently translated these contents back into English. Subsequently, the two researchers compared two English versions and jointly revised the first Chinese version of the project. Then we invited a panel of experts in the field of consumer behavior to test the surface validity of the survey tools. Based on their feedback, we have made minor modifications to improve the comprehensiveness and user-friendliness of the measurement project. The measurement items of the questionnaire design were tested using a seven-point Likert scale, which ranged from 1 (not agree at all) to 7 (absolutely agree).

Data collection

Although the current popularity of short videos in China is very high, young people are more likely to view short videos. Therefore, this study is mainly survey research, carried out in Chinese universities. First, based on its purpose, this study did not invite subjects lacking familiarity with short video platforms to participate in the survey; thus, the final sample has good validity. Second, the surveyed students completed their questionnaires during class to ensure the highest quality thereof. Finally, to help them understand the items more clearly, this study collected data through face-to-face structured interviews with the respondents rather than allowing them to complete them on their own. The average completion time of a single questionnaire was approximately 18 min.

Data collection was conducted during the summer of 2022. A total of 400 responses were received over a four-week period. Therefore, 350 responses were used for subsequent analysis, after 50 incomplete and invalid responses were dropped. Table 1 summarizes the demographic information of the final sample. As Table 3 indicates, among the 350 respondents, 98 were male, and 252 were female. There were significantly more women than men who watched the short videos. According to the <Tiktok Research Report> released by Beijing News, female users account for 65.4%, which is much higher than the 34.6% of male users (Tiktok Research Report, 2023), which is consistent with the results of this study. This is because short video content is relatively easy and entertaining, as well as more intuitive and convenient shopping channels, attracting more female users. The age group of 21 to 29 was the largest, containing over half of the respondents, or 53.7%. Approximately 94.9% of the respondents had a monthly income of less than RMB 3000 (equivalent to US $500). Concerning short video viewing frequency, 163 participants stated they view short videos for less than 1 h a day.

Table 3 Demographics of the survey respondents (N = 350).
Full size table

Data analyses and results

Common method bias

To minimize the impact of CMV, this study adopted control measures such as anonymous filling, setting antisense items, and disrupting the order of items. Harman’s single-factor method was also used to test for common method bias in the research. These results show that the measurement items are all aggregated into 10 factors with eigenvalues greater than 1 and that the first factor explains 28.587% of the variance in all items, which is less than the critical value standard of 40%, indicating that this study effectively controlled for data variance and thus, CMV.

Reliability and validity

The Cronbach’s α and composite reliability (CR) values were above the recommended threshold of 0.7 (Straub et al., 2004) and exhibited a satisfactory level of reliability (as in Table 2). In Table 4, this study verified the convergent validity using the average variance extracted (AVE) and factor loadings. The results show that the AVE value is higher than the recommended threshold of 0.5 (Fornell and Larcker, 1981) and the standard loadings are higher than 0.7, thus indicating good convergent validity (Chin et al., 1997).

Table 4 Results of confirmatory factor analysis.
Full size table

By comparing the correlation coefficients with the square roots of AVE (Fornell and Larcker, 1981), the study found that all square roots of AVE were greater than the interconstruct correlation coefficients. Therefore, the discriminant validity of the study is good (as shown in Table 5).

Table 5 Results of correlation coefficient testing.
Full size table

Hypothesis testing

After examining the measurements’ validity and reliability, we tested the proposed hypotheses using AMOS. Table 6 and Fig. 3 indicate that nine of the hypothesized relationships were supported by these results. Credibility significantly influenced shared value creation, thus supporting H1 (β = 0.122, p < 0.01). Consistency significantly influenced shared value creation, thus supporting H2 (β = 0.152, p < 0.001). Completeness significantly influenced shared value creation, supporting H3 (β = 0.121, p < 0.05). Clarity significantly influenced shared value creation, supporting H4 (β = 0.213, p < 0.001). Likability significantly influenced shared value creation, supporting H5 (β = 0.105, p < 0.05). Empathy significantly influenced shared value creation, supporting H6 (β = 0.113, p < 0.05). Similarity significantly influenced shared value creation, thus supporting H7 (β = 0.072, p < 0.05). Similarity significantly influenced shared value creation, thus supporting H8 (β = 0.293, p < 0.001). In addition, shared value creation significantly influenced purchase intention on short video platforms, thus supporting H9 (β = 0.878, p < 0.001).

Table 6 Results of hypothesis testing.
Full size table
Fig. 3: The results of the research model.
figure 3

*p < 0.05, **p < 0.01, ***p < 0.001.

Full size image

Mediating effect testing

Since this study suggests that shared value creation mediates the effect of persona perception on purchase intention on short video platforms, bootstrapping was used to test this mediating effect (Preacher and Hayes, 2008, Shrout and Bolger, 2002). The use and testing of mediating effects is a main trend in management studies. SEM based on bootstrapping can overcome the shortcomings of traditional testing methods such as the Sobel test for small sample sizes and mediating effect values. Table 7 therefore shows that the indirect effect of shared value creation on the relationship between persona perception and purchase intention on short video platforms is significant, with a 95% bootstrap confidence interval, excluding zero. This finding suggests that shared value creation mediates the effect of persona perception on purchase intention on short video platforms.

Table 7 Results of mediating effect analysis.
Full size table

Moderation effect testing

To verify Hypothesis 10, that is, the moderating effect of regulatory focus on shared value creation and purchase intention, we first set purchase intention as the dependent variable and then located shared value creation, regulatory focus, and the interaction term between them in turn. To eliminate the collinearity problem, shared value creations and regulatory focus should be centrally processed, and then the interaction term between these two variables should be constructed. These results show that the interaction term between shared value creation and regulatory focus is significantly positively correlated with purchase intention (β = 18.717, p < 0.01), which indicates that regulatory focus has a moderating effect on shared value creation and purchase intention (see Fig. 4). Furthermore, in the measurement of regulatory focus in the research design, the 11 items of Higgins et al. (2001) were used to measure consumers’ regulatory focus attributes, and consumers were classified according to the measurement results. In the regulatory focus questionnaire, questions 1–6 measure promotion focus, and questions 7–11 measure prevention focus. In the classification method of this research, the difference between promotion focus and prevention focus is set as M; with M = 0 as the reference line, M greater than 0 is classified as promotional focus, and M less than or equal to 0 is classified as prevention focus. Depicting these differences among different individuals in the relationship between shared value creation and purchase intention also proves that regulatory focus plays a moderating role between shared value creation and purchase intention (see Table 8). Thus, Hypothesis 10 is verified.

Fig. 4: Research framework and empirical analysis.
figure 4

The moderating effect of regulatory focus.

Full size image
Table 8 Results of regulatory focus moderating effect analysis.
Full size table

To verify Hypothesis 11, the same method was used to centralize shared value creation and social presence to construct an interaction term between these two variables. These results show that the interaction item between shared value creation and social presence is significantly positively correlated with purchase intention (β = 11.470, p < 0.05), which indicates that the greater the level of social presence is, the stronger the positive effect of shared value creation on purchase intention (see Fig. 5). Furthermore, social presence is divided into two samples (high and low) by adding or subtracting one standard deviation from the mean value of social presence and depicting the differences in the relationship between shared value creation and purchase intention among individuals with different levels of social presence. These results prove that social presence positively regulates the impact of shared value creation on purchase intention. However, the low social presence group demonstrates no significant moderating effect between shared value creation and purchase intention (see Table 9). Thus, Hypothesis 11 is verified.

Fig. 5: Research conclusions and marketing implications.
figure 5

The moderating effect of social presence.

Full size image
Table 9 Results of social presence moderating effect analysis.
Full size table

Robustness check

To enhance the strength of the research conclusions, this study used PROCESS in SPSS 24.0 to analyze the robustness of the research model. Model 14 was used to test the moderated mediation effect of the model and indicated that the 95% confidence interval of Bootstrap contained zero. However, the mediating effect of shared value creation and the moderating effect between shared value creation and purchase intention was shown to be significant. These therefore fully proved the reliability and robustness of the hypothesis test results in this study.

Discussion and implications

Discussion of findings

This study introduces the “S-O-R” model into the short video marketing context, focuses on the antecedent, mediator, and moderator variables of consumption intention, and explores the relationship among persona perception, shared value creation, regulatory focus, social presence, and purchase intention.

First, in short, video marketing, the perception of each dimension of persona has a significant positive impact on shared value creation, consistent with previous research (Salminen et al., 2020). According to the findings, willingness to use has the greatest impact on shared value creation. There are various persona positions in current short video marketing, and those that are more suitable for short video users create more shared value for both parties.

Second, shared value creation plays a mediating role between persona perception and purchase intention and has a significant positive impact on purchase intention. Moreover, shared value creation has a very strong impact on purchase intention. The research on shared value creation referenced (Ham et al., 2020) in this study has mainly discussed the economic value and social value provided by short videos. The short video marketing model extends the platform value chain, which is conducive to the integration of short video and e-commerce platforms and the creation of stakeholder value, i.e., shared value creation.

Finally, regulatory focus and social presence play a moderating role between shared value creation and purchase intention. Specifically, when shared value creation reflects positive results, the purchase intention of promotion-focused consumers is stronger than that of prevention-focused consumers. When shared value creation reflects negative results that must be avoided, the purchase intention of prevention-focused consumers is stronger than that of promotion-focused consumers. In addition, amid high shared value creation, users experiencing a greater social presence have a stronger purchase intention, consistent with previous research (Algharabat et al., 2018). In contrast, amid low shared value creation, social presence has no significant moderating effect on purchase intention. This may be because in this context users watch short videos for a very short time. If the social presence they experience is low, users directly scroll to the next short video, making it difficult for them to form a purchase intention.

Implications

First, at the managerial level, the managers should attach great importance to short video marketing as an effective means, as an important tool to promote sales growth and brand building. Our finding shows that optimizing short video anchors and content can enhance users’ persona perception and shared values, thereby increasing customers’ purchase intention. As a brand new form of media, firms can elaborate and release various types of marketing videos around their own products and services, transform the value of short videos into the value of specific brands, and produce an effective brand fission effect (Wang et al., 2023). At the same time, firms need to develop reasonable short video marketing plans, including target customers, marketing strategy, budget, and other aspects of planning (Dong et al., 2023).

Second, at the social level, firms should create a social presence for users through short video marketing. This study documents that a user’s social presence is closely related to purchase intention. Firms can realize users’ emotional release and entertainment needs through situational, funny, and personified soft advertising. In addition, firms can also use virtual technologies such as interactive artificial intelligence and mixed reality to provide users with a more authentic and pleasant shopping experience (Sung et al., 2021). In the context of social e-commerce, firms can also carry out social activities such as live-streaming through short video platforms to enhance users’ social presence (You et al., 2023).

Third, at the economic level, firms can improve the exposure and visibility of products through short video marketing, expand market share, increase sales, and thus obtain more commercial value. At the same time, firms need to formulate reasonable price strategies and promotion programs according to market conditions and user needs and pay attention to the integration and coordination with other marketing channels. In addition, firms can reduce costs, improve efficiency and enhance their competitiveness through short video marketing (Dai and Wang, 2023).

Finally, at the level of consumer behavior, consumers need to spend a lot of time and energy to find the information that meets their needs and hobbies in the mass information of short video platforms. Therefore, personalized functions can meet the needs of this audience. Brands and merchants can use the Internet and big data technology to recommend products or services with high similarity based on user profiles and recommend short video content that users are interested in through personalized algorithms (Xiao et al., 2023). In addition, firms can use the shopping cart function of the short video platform to package and sell multiple goods, carry out group buying, and other activities to attract consumers to buy at preferential prices. The short live video function is used to realize experiential consumption scenarios such as product trial and fitting, thereby enhancing consumers’ purchase desire and ultimately promoting the completion of transactions (Wang et al., 2022).

Limitations and future research

The limitations of this study should be accounted for before generalizing its findings. First, this research uses cross-sectional data to measure consumers’ cognition of short videos at a certain time, but data from different periods can have further implications for research. Future research could thus expand the scope of this study, increase the sample size, and collect more comprehensive data on live broadcast consumers to improve the external validity of the above findings. Secondly, the results of this article may have application limitations in different regions and cultures. This study was conducted in China, and due to the different psychological states of consumers in different countries and cultures, it may limit the applicability of its results in other cultures (Sun and Zhang, 2006). Therefore, it is crucial to conduct cross-regional and cross-cultural research in the future. Finally, although this study considers the relationship among the independent variables, mediator variables, and moderator variables, due to the ongoing development of short video marketing in China, new research phenomena continue to emerge. Hence, future research should also consider variables from other perspectives, such as the product perspective, using product involvement as a moderator variable.