Study on the extension path of agricultural products live marketing mode of rural e-commerce platform based on network traffic analysis
Published Online: Feb 03, 2025
Received: Sep 09, 2024
Accepted: Dec 29, 2024
DOI: https://doi.org/10.2478/amns-2025-0021
Keywords
© 2025 Zhengxi Zhou et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
With the rapid development of information technology, the integration of the Internet and agriculture has become an irreversible trend. As an innovative marketing means, live marketing of agricultural products e-commerce has been loved by consumers for its convenient, efficient and intuitive features [1–2]. Through e-commerce live broadcasting, agricultural products can break through the limitations of time and space, realising the zero-distance communication between producers and consumers, thus effectively promoting the sales and promotion of agricultural products. However, in the context of rural revitalisation, agricultural product’s e-commerce live marketing strategy faces many challenges and problems [3–4].
The first task of live marketing is to provide high-quality and attractive content. The content should be unique, able to arouse the interest of the audience and satisfy their needs [5–6]. Whether it is a product demonstration, industry insight or an entertaining event, the creativity and usefulness of the content is the cornerstone of successful live marketing [7]. Another important feature of live marketing is real-time interaction. Viewers are able to ask questions and make comments during the live broadcast, while the anchor can respond instantly [8]. This interactivity increases the engagement of the live broadcast and enhances the intimacy between brands and consumers. Reasonable use of features such as pop-ups and likes motivates audience participation and further leads to interaction [9]. The success of live streaming marketing lies not only in the live streaming process itself but also in the social sharing of viewers. By providing easy-to-share content and interactive sessions, brands can leverage viewers’ social networks to quickly diffuse information and achieve word-of-mouth [10]. The integration of social media platforms is also an important means to further increase brand exposure through cross-platform promotion [11].
At the current stage, China’s digital economy is gradually developing upstream and downstream penetration of the supply chain of various industries. The penetration rate of China’s live e-commerce industry was 4.3% in 2019, an increase of 4.03% from 0.27% in 2017, and China’s live e-commerce penetration rate of various industries reached 9.3% in 2020 [12–13]. The digital supply chain has become an opportunity for the integration and development of various industries and information technology, with the help of big data, artificial intelligence and other technologies, to substantially improve the circulation efficiency of the industrial chain and the level of external services, while the impact of live streaming with goods on traditional agriculture is more significant [14]. Including the reduction of agricultural e-commerce circulation links, reduced circulation costs, but also innovative agricultural products e-commerce marketing model for agricultural e-commerce to inject a new impetus and actively help rural e-commerce to achieve leaps and bounds in the development of the countryside [15].
Live streaming traffic is one of the determining factors for the sales of e-commerce enterprises, so exploring its impact on the traffic can help managers of agricultural enterprises to start from the service scene of the live streaming room, under the support of the service scene theory and the 4Cs theory, and from the enterprise’s point of view, catering to the needs of customers to watch and innovate the form of the live streaming, and combining the characteristics of different agricultural products to be sold to customise the live streaming sales mode that meets their own needs and meets the needs of consumers. To meet the needs of consumers, achieve more efficient operation, and promote the sustainable and healthy development of the industry. Literature [16] combines quantitative and qualitative analysis methods, analyses the FACEBOOK index of live sellers and the live sales process, and confirms a variety of scientific sales methods and sales strategies. Literature [17] examined the current situation of live broadcast marketing and the practice of marketing theory in the context of the Internet and concluded that live broadcast marketing needs to constantly innovate in content and form, followed by the reasonable use of information technology similar to VR and other information technology to improve the effect of live broadcasting, and at the same time, integrating multi-channels in order to create intelligent live broadcast marketing. Literature [18] discusses the development history of social media and the role of live video deeply analyses the strategy of live video, as well as assesses the impact of live video. Literature [19] categorised live streaming interactions and built an analytical framework to examine the impact of live streaming interactions on consumption and marketing, and the study pointed out that IPI and IPD interactions drove subservient consumption and socialisation, and enhanced consumers’ purchase intentions, whereas IMI interactions only promoted social presence. Literature [20] examined the feasibility of a live corporate marketing approach using qualitative content analysis methods and interviews, and the results of the study suggest that going live to a corporation is an effective way of branding that will have more potential in the future. Literature [21] examined the attitudes of flower and tree family farms towards live marketing. Based on structural equation modelling, it was found that young flower and tree farm operators were more receptive to live marketing, and it was also pointed out that factors such as subjective knowledge, governmental support, and perceived behavioural control have a diminishing effect on the attitudes of farm owners towards live marketing. Literature [22], based on SWOT analysis arguments, conceived the farmers’ live agricultural products operation strategy that combines the principle of minimising constraints + maximising product value, which can build practical live agricultural products for farmers.
With the continuous development of society, all walks of life towards the direction of industry informationization is not a new development. Industry development space has not only been broadened, “live + e-commerce” sales model is gradually becoming the first choice of most agricultural products to reduce sales costs and improve service quality. The creation of this approach allows farmers, customers and merchants to achieve efficient communication, so the establishment of a “live + e-commerce” sales system to try to meet the supply side and demand side, supply and demand, supply and demand for agricultural products to maximize the satisfaction of the needs of the seller and the buyer is an issue that deserves attention. This is an issue that deserves attention. Based on the stimulus-organisation-response theoretical framework, literature [23] provides an in-depth analysis of how the features of agricultural live streaming affect consumers’ willingness to buy and green consumption perceptions, and the study points out that influences, promotions, and interactive entertainment elements positively affect consumers’ willingness to buy. Literature [24] built a live e-commerce assessment framework on the basis of live streaming research. It combined the questionnaire survey method and structural equation modelling, which confirmed that the quality of e-commerce live streaming positively affects the green trust of users in terms of five dimensions such as the quality of information, system quality, quality of service and social presence, etc. The study provides an important reference for enterprises to improve the quality of live streaming. Literature [25] discusses the factors influencing consumers’ purchase intention and purchase attitude in agricultural products public welfare live broadcasting around the three dimensions of platform, product and consumer, and the study reveals that the perceived interactivity of the agricultural products live broadcasting process, perceived endorsement, and familiarity with the products positively affect the purchase intention of agricultural products in the consumer dispute. Literature [26] examined the nature of cross-border agricultural supply chain in the context of the rapid development of the live broadcasting industry and, based on the rooting theory, platform theory, and ecosystem theory, constructed a cross-border supply chain influencing factors analysis model for agricultural products, which helped stakeholders to improve their knowledge of cross-border supply chain for agricultural products. Literature [27] empirically confirmed the impact of short video live streaming on the sales of agricultural products and that the charisma of the leader in the live streaming process, the richness of the live streaming content, and the evaluation of the live streaming interaction all influence the customer’s cognition and emotion, which then affects the consumer’s willingness to purchase agricultural products. Literature [28], in a study on the characteristics of live broadcasting and sales of agricultural products, pointed out that nearly 40% of consumers were dissatisfied with the purchase of agricultural products through live broadcasting and pointed out that consumers’ willingness to repeat the purchase of agricultural products is affected by the quality of information, as well as the satisfaction of the purchase of agricultural products through live broadcasting. Literature [29] used expectation theory and SERVQUAL framework to try to study the customer consumption influencing factors from product live streaming as well as strategies to drive consumers’ agricultural product purchases, and the study elucidated consumers’ higher satisfaction with live streaming of agricultural products, in which the live streaming platform has the strongest direct influence on customer satisfaction, and the category of agricultural products has the strongest indirect and comprehensive influence.
This paper takes consumer viewing experience as an explanatory variable, network traffic analysis as a mediator variable, and live broadcast mode as a moderating variable, in order to explore the consumer purchasing behaviour under the live broadcast marketing mode of agricultural products on rural e-commerce platforms based on network traffic analysis. Five regression models were constructed to verify the influence of consumer viewing experience on consumer purchasing behavior, the role of network traffic analysis, and the moderating role of rural e-commerce live marketing mode. Finally, 933 rural e-commerce platform merchant accounts are used as research samples to validate the theoretical models in the previous paper through the main effect test, mediating role test, and moderating effect test.
Consumer Viewing Experience and Consumer Purchase Behavior
Consumer viewing experience can enhance consumers’ perception of safety and purchase attitude in virtual shopping centers (online live shopping platforms), and it is even a prerequisite for consumers to use virtual shopping centers. In the process of rural e-commerce live broadcasting, the consumer’s non-textual language clues are ignored and can only convey textual, anthropomorphic attributes, which will be conveyed to the anchor of the relevant information to like, comment and other textual behavior “anthropomorphic” feedback, a greater number of comments and likes and other relevant information will be in the form of All relevant information such as more comments and likes will be instantly reflected in the live broadcast in the form of “pop-up”, so that the whole audience can receive relevant information, resulting in a strong consumer viewing experience [30]. This creates a positive buying attitude and makes real purchases. As a result, the following hypotheses are proposed in this paper:
H1: Consumer viewing experience positively influences consumer purchase behavior.
The mechanism of consumer viewing experience influencing purchase behavior: the role of network traffic analysis
Consumers, out of trust in the platform and the anchor, will stay in the live broadcast for a long time in order to understand the goods displayed by the anchor and purchase the live broadcast goods for a long time. It can be seen that the network traffic analysis can clarify the consumer’s stay time in the live room, the number of times in and out, the frequency of interaction, etc., so that the marketing strategy can be adjusted in a timely manner. The increase in viewing time means that the probability of consumers generating purchasing behavior will increase, and long-term stable consumer viewing can increase the repurchase rate of goods brought by the live room. As a result, this paper proposes the following hypothesis:
H2: Network traffic analysis plays a mediating role in the influence of consumer viewing experience on their purchasing behavior.
The moderating role of rural e-commerce live marketing model
With the continuous development of the rural e-commerce live marketing model, the industry has gradually formed two new rural e-commerce live marketing models: the anchor live model and the brand live model. The first model is a single person multi-brand model, which is mainly manifested in the anchor with goods marketing model. The main feature of this model is that the anchor is dominated by a shorter live time. Another live mode for a multi-person single brand marketing model. These anchors are used during the live broadcast process to achieve the goal of changing people’s behavior by broadcasting continuously for long periods of time. In the “anchor live” mode, the anchor is a live commodity displayer, intuitive feedback to consumers on the content of the commodity, and real-time communication with consumers, which can mobilize consumer emotions and become a user group topic guide. They have the role of influencing user perception in the process of rural e-commerce anchor marketing, can significantly guide the tendency of public opinion of marketing, and positively attract consumers to watch their live broadcasts, thus increasing the stay time of consumers in the live broadcast room. As a result, this paper proposes the following hypotheses:
H3a: The anchor live broadcast mode positively regulates the relationship between consumer viewing experience and network traffic analysis.
“Brand live” live with goods, the use of replacement without broadcasting behavior, different live anchor style and clinical performance has a large difference, the user continuity of behavior is interrupted may produce negative cognitive reflection, and a long time of live broadcasting will make consumers produce aesthetic fatigue, but also reduces their willingness to watch. In this way, the brand live broadcast through the personnel change so that the user’s coherent viewing experience is blocked, and the viewing time becomes shorter. Consumers watch live brand broadcasts mainly out of high recognition of the brand, but their viewing time and engagement with the broadcasts will be reduced. As a result, this paper proposes the following hypothesis:
H3b: The brand live streaming model negatively regulates the relationship between consumer viewing experience and web traffic analysis.
Based on the above analysis, the theoretical model formed is shown in Figure 1. The explanatory variables, mediating variables, moderating variables, and explained variables are consumer viewing experience, network traffic analysis, live broadcast mode, and consumer purchase behavior, respectively.
Theoretical model
Explanation of Variables
Explained variable: consumer buying behaviour (BU). Since the quantitative subject of measuring consumer behavior in this paper is the anchor information, for this reason, the average produce sales volume of the anchor field is adopted as the quantitative index of consumer buying behavior.
Explanatory variable: consumer viewing experience (CVE). According to the previous theoretical analysis, consumer presence is mainly manifested in the psychological perception of interactive response, emotional response and cohesive response of participants under the social interaction of network media. Specifically, this paper takes the sum of the standardised data of the four items related to the number of likes (THU), the number of comments (DEM), the length of the live broadcast (TIM), and the number of live broadcasts (NUM) as a quantitative way to quantify the consumer viewing experience.
Mediating variable: network traffic analysis (NFA). Consumers’ network traffic for rural e-commerce live streaming bandwagon can be measured from two aspects: the number of fans and the number of repeated viewings, i.e., network traffic = the number of times entering the live streaming room (STU) / the number of people entering the live streaming room (SWN) + the number of fans (FNU) / number of people watching (WNU) [31].
Moderating variable: live broadcast mode (MDO). In this paper, the dummy variable is set by manual retrieval for the live broadcast mode, and the anchor live broadcast is assigned as 1, while the brand live broadcast is 0.
control variables: control variables include the number of times consumers enter the shop where the goods are located (NUM), the number of people who enter the shop (PEO), the price per unit of goods sold (PRI) and the rate of increase in the number of fans (FAN), and the length of time of the live broadcast (TIM).
Model construction
In order to verify the influence of consumer viewing experience on consumer purchasing behaviour in H1, a multiple linear regression model is constructed:
To verify the mediating role of web traffic analysis in H2 in consumer viewing experience and consumer purchase, the following multiple linear regression model was constructed:
In order to verify the moderating effect of different live streaming modes in H3a and H3b on consumer viewing experience and network traffic analysis, the cross-multiplication term between live streaming modes and consumer viewing experience is added, and the following model is constructed:
Where
Data sources
The study selects 933 rural e-commerce merchant accounts on a platform. Statistics on total sales, number of likes, number of comments, length of live broadcast, number of live broadcasts, number of live broadcasts, number of fans, unit price of goods sold, and length of time on air for each live broadcast from 1 April 2023 to 28 August 2023 are presented.
The descriptive statistics of the variables are shown in Table 1. The empirical analysis was conducted on 933 rural e-commerce platform merchant accounts. The results of the descriptive statistics show that there are significant gaps in sales, likes, comments and other variables of rural e-commerce platform merchant accounts. For example, the difference between the maximum and minimum values of the anchor field average agricultural sales (BU), the number of likes (THU), and the number of comments (DEM) are 175999867, 368989764, and 168755388, respectively, which indicates that the sample data used in this paper are well represented.
Descriptive statistics of variables
| Variable | Sample size | Min | Max | Mean | SD | Median |
|---|---|---|---|---|---|---|
| BU | 933 | 133 | 176000000 | 1936584 | 8639121 | 269048 |
| THU | 933 | 10236 | 369000000 | 1987463 | 2469521 | 185624 |
| DEM | 933 | 10064 | 168765452 | 989654 | 1469500 | 176321 |
| STU | 933 | 44 | 20010000 | 196523 | 196857 | 43655 |
| SWN | 933 | 630 | 5660000 | 66854 | 25631 | 206963 |
| FNU | 933 | 633 | 5022220 | 25632 | 52631 | 204589 |
| WNU | 933 | 16 | 933000000 | 19600000 | 9652000 | 396660 |
| MDO | 933 | 0 | 1 | 0.66 | 0.56 | 0.46 |
| NUM | 933 | 6 | 18963 | 9632 | 763 | 2364 |
| PEO | 933 | 3 | 1866 | 963 | 89 | 685 |
| PRI | 933 | 3.5 | 17000 | 406.9 | 63.8 | 186 |
| FAN | 933 | 0 | 0.5624 | 0.0196 | 0.0016 | 0.0109 |
| TIM | 933 | 1.55 | 176.36 | 7.636 | 25.33 | 6.08 |
The above data was standardised, and the standardised results were used for correlation analysis. The main variables correlation detection is shown in Figure 2. The correlation between the main variables of consumer purchase behaviour (BU), the number of likes (THU), the number of viewers (WNU), the network traffic analysis (NFA), and the mode of live broadcasting (MDO) are all significant with p<0.01, passing the correlation test.
Major variable correlation detection
The results of the main effects test are shown in Table 2. Where *, **, *** denote p<0.05, p<0.01, p<0.001 respectively (the same below). Model (1) indicates that consumer viewing experience has a significant positive effect on consumer purchasing behaviour, regression coefficient = 1.832, p<0.05, hypothesis H1 is valid.
Main effect test results
| Variable | CVE | NFA | MDO | NUM | PEO | PRI | |
|---|---|---|---|---|---|---|---|
| Model(1) | 1.832* | - | - | -0.236*** | 0.367*** | -0.016 | |
| t | 7.365 | - | - | 6.354 | 7.325 | 5.264 | |
| Variable | TIM | FAN | CVE*MDO | R2 | DR2 | F | |
| 0.035 | -0.027 | - | 0.736 | 0.044 | 6.05 | ||
| t | 5.632 | 6.254 | - |
For the control variables, only the number of times consumers enter the shop where the product is located, and the number of people entering the shop present a significant effect on purchasing behaviour in the test results. Specifically, the higher the number of times consumers enter the shop where the goods are located and the number of people entering the shop, the stronger the purchasing behaviour towards the agricultural products displayed in the live broadcast.
The results of the mediating role test are shown in Table 3. Models (2) and (3) show that web traffic analysis has a positive contribution to consumer purchasing behaviour, with regression coefficient = 6.325, p<0.05. The above results confirm that web traffic analysis plays a mediating role in influencing consumer purchasing behaviour, and Hypothesis 2 is verified.
The mediation action test results
| Variable | CVE | NFA | MDO | NUM | PEO | PRI | |
|---|---|---|---|---|---|---|---|
| Model(2) | 3.106*** | - | - | 8.639 | -6.235 | -2.365 | |
| t | 20.337 | - | - | 20.635 | 17.635 | 16.355 | |
| Variable | TIM | FAN | CVE*MDO | R2 | DR2 | F | |
| -3.256 | 6.595* | - | 0.736 | 0.114 | 12.33 | ||
| t | 16.325 | 20.365 | - | ||||
| Model(3) | Variable | CVE | NFA | MDO | NUM | PEO | PRI |
| 3.265*** | 6.325* | - | -0.135* | -0.015* | 0.355*** | ||
| t | 19.635 | 18.265 | - | 16.325 | 17.336 | 20.584 | |
| Variable | TIM | FAN | CVE*MDO | R2 | DR2 | F | |
| -0.044 | -0.018 | - | 0.735 | 0.147 | 17.33 | ||
| t | 16.335 | 17.258 | - |
The results of the moderating effect test are shown in Table 4. In this study, we use consumer purchase behavior as the dependent variable, anchor marketing mode, brand marketing mode, and their interaction terms as the independent variables, and center each question term on the interaction term. According to model (4), it can be seen that the coefficient of the interaction term of the anchor marketing model has increased compared to the results of model (1) when explaining consumer purchase behaviour, with a regression coefficient = 1.963, p<0.05, and the R2 of the model (4) has increased by 0.8% compared to model (1). Therefore, the anchor marketing model positively moderates the relationship between consumer viewing experience and web traffic analysis, and hypothesis 3a is verified. According to the results of model (5), its interaction term coefficient is significantly negative after adding adjustment variables, indicating that the brand marketing model negatively regulates the effect of user perception on user stickiness, and H3b is verified.
Regulation effect test results
| Variable | CVE | NFA | MDO | NUM | PEO | PRI | |
|---|---|---|---|---|---|---|---|
| Model(4) | 0.077** | - | 1.965*** | 0.086 | -0.076 | -0.019 | |
| t | 6.532 | - | 6.324 | 6.324 | 8.523 | 6.254 | |
| Variable | TIM | FAN | CVE*MDO | R2 | DR2 | F | |
| 0.011 | 0.035 | 1.963* | 0.744 | 0.149 | 13.66 | ||
| t | 6.254 | 8.562 | 15.631 | ||||
| Model(5) | Variable | CVE | NFA | MDO | NUM | PEO | PRI |
| 0.027*** | - | -0.347* | -0.011 | 0.025 | 0.577 | ||
| t | 15.664 | - | 6.324 | 6.254 | 8.654 | 6.441 | |
| Variable | TIM | FAN | CVE*MDO | R2 | DR2 | F | |
| -0.119* | -0.055 | -0.328*** | 7.66 | 0.147 | 13.66 | ||
| t | 6.512 | 6.325 | 8.569 |
In this study, the model was subjected to the main effect test, moderating effect test and mediating effect test as mentioned above, and after a detailed statistical analysis and testing process, the results of all the research hypotheses were obtained, as shown in Table 5. All four hypotheses proposed in this paper are valid. It shows that the live marketing model of agricultural products on rural e-commerce platforms based on network traffic analysis can promote consumer purchasing behavior.
The model assumes the test results
| Research hypothesis | Whether or not |
|---|---|
| H1:Consumer watching experience is affecting consumer buying behavior | Yes |
| H2:Network traffic analysis is mediated in the influence of consumer watching experience on its purchase behavior | Yes |
| H3a:The broadcast mode of the anchor is adjusting the relationship between the consumer viewing experience and the network traffic analysis | Yes |
| H3b:The relationship between the user viewing experience and the network traffic analysis is adjusted for the brand live mode | Yes |
Improve basic network facilities and safeguard the logistics system
Optimise the mobile network infrastructure live agricultural products marketing to help rural revitalisation, continue to promote the Internet penetration rate in key villages, and improve the quality of rural Internet users’ access to the Internet. So that more rural residents can successfully access the network and be comfortable with live agricultural product sales, increase efforts to improve the basic network facilities, and lay a good foundation for villagers to carry out online live marketing [32].
Optimise the logistics and distribution system. After the live marketing of agricultural products is carried out, it is necessary to rely on logistics to distribute goods to consumers. Compared with the city, the logistics and distribution system in rural areas is more dispersed, and it is not easy to form an intensive economic logistics effect based on the village, which requires the government to invest in infrastructure construction. Combined with the layout of local villages and towns, scientific planning of transport routes, the construction of logistics distribution centers, warehousing centers, and logistics distribution channels. For the circulation of vegetables and fruits, fresh agricultural products and pre-prepared dishes, it is also necessary to build a standardised chilled transport system and invest in the construction of cold storage, cold cabinets and other facilities and equipment. The relative cost of independent investment is relatively high, and cold chain logistics cooperation can be achieved with professional logistics companies to provide the necessary subsidies.
Rural e-commerce talent recruitment, with the help of local colleges and universities
policy guarantee “new farmers” return home to help rural revitalization needs talent as a guarantee, led by the government, the introduction of relevant policies, the organization of agricultural technology training and digital technology training, to help townships and cities new young people to return to their hometowns to start their businesses, professional and technical personnel and college students to return to their hometowns to do village cadres and other help to revitalize the countryside. The government has introduced policies to support the “Three Rural Issues”, increased subsidies and assistance, reduced redundant approval processes, and effectively provided benefits to “new farmers”, “new agricultural organizations” and “new agricultural enterprises”. New farmers are equipped with stronger information technology means to carry out short video operations and rural e-commerce live operations. To “help - help - bring” way to drive rural agricultural sales and rural tourism project development.
the power of joint colleges and universities to build a brand can take advantage of the power of the local universities, led by the government, and work together to build a “government-school-village” synergistic platform. Colleges and universities to provide professional rural e-commerce talents, towns and villages to provide high-quality agricultural products, and jointly build on- and off-campus live practice bases. Agricultural production bases and processing sites can be used as places for college students to practice short video filming and rural e-commerce operations. On-campus teaching with a live audience can be used as the live sales venue for the township. This could compensate for the shortcomings of the lack of talent in the townships and, at the same time, aid in the development of exceptional rural e-commerce talents.
Joint Agricultural Chamber of Commerce and Leading Enterprises to Help Rural Transformation and Upgrading
Leveraging the Agricultural Business Association township agricultural sales promotion, agricultural enterprises play a certain role. Agribusiness Association as an organizational structure, gathered the local agricultural company has considerable practical experience, while the enterprise has no shortage of young entrepreneurs, receiving new rural e-commerce concepts and the ability to live marketing skills will be much stronger. In addition, in addition to its land operation, agriculture will take the way of contracted land to grow agricultural products. These types of villages and towns tend to have richer agricultural products, which is very useful for enhancing the local production value.
Helping rural transformation rural revitalization is to develop the industry of the countryside and make the countryside more affluent. Agribusiness in rural areas to develop their specific agricultural products industry, the same can be led by the government to guide the construction of agricultural products industry belt, so that farmers participate in the upstream and downstream of the production and processing of agricultural products. Villagers can use a richer way to carry out live sales, as the owner has more autonomy and a sense of belonging can play the marketing creative point will be more. The government, as the leading role, should formulate appropriate policies to provide technical services to agricultural companies, enterprises involved in rural revitalization, and revitalization of agricultural products industry belt.
Cultivate famous agricultural products, excellent talents and famous villages, and form characteristic brands
Live marketing, as a form of new rural e-commerce, only changes the shopping scene, but the content also needs to have the core keywords of the category and form a logo on the system. Attention should be paid to cultivating the brand IP awareness view of farmers and labeling their accounts in the minds of consumers. Farmers should be encouraged to learn how to imitate other accounts to build their brand, how to package the account, and shape the distinctive brand matrix. Building a distinctive brand can focus on shaping the marketing strategy from three perspectives: cultivating famous agricultural products, excellent talents, and famous villages. Famous agricultural products have higher product standards and a larger audience, which further enhances the brand image. As a key opinion leader, excellent talents have a certain appeal among fans in specific fields and can act as the head anchor to create an account matrix. Famous villages have typical marketing points, and it is worthwhile for villagers to work together to build a characteristic geographical identity.
Based on the theory that network traffic analysis can enhance consumers’ viewing experience, this paper constructs a theoretical model of the influence of consumers’ viewing experience on purchasing behaviour, and conducts regression analyses with network traffic analysis as a mediating variable and live broadcasting mode as a moderating variable respectively.
The study found that, firstly, consumer viewing experience significantly and positively influences consumer purchasing behaviour. Second, network traffic analysis plays a mediating role in the influence path of consumer viewing experience on consumer purchasing behaviour, i.e., network traffic analysis can significantly enhance consumer viewing experience, which in turn enhances consumer purchasing behaviour. Third, different live broadcast modes influenced the relationship between consumer viewing experience and web traffic analysis. The anchor mode positively regulates the relationship between consumer viewing experience and web traffic analysis. The brand’s live broadcast mode negatively regulates the relationship between consumer viewing experience and network traffic analysis. Accordingly, this paper proposes to strengthen the basic network facilities, leverage the power of local universities, unite agricultural chambers of commerce and leading enterprises, and cultivate featured brands to expand the live marketing mode of agricultural products on rural e-commerce platforms based on network traffic analysis.