Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 20, No. 2, 2025 42 Content Marketing Strategy and User Behavior Analysis of Short Video and Live E-commerce Jundong Qian Three's Company Media Group Co., Ltd. Beijing, China gongzuo1232024@126.com Abstract: In this paper, a three-stage content marketing model of "attraction-conversion-repurchase" is constructed, integrating scenario-based content construction, KOL/KOC synergy mechanism, real-time interaction and personalized recommendation strategy. The model is based on user behavior path and content response mechanism, combined with collaborative filtering and behavior scoring technology. The experiment selects the cross-platform marketing data of a national beauty brand in Shake, Taobao live and Xiaohongshu platforms for validation. The results show that the click-through rate in the Shake Shine attraction phase is 8.9%, the conversion rate of Taobao live broadcast reaches 9.6%, and the repurchase rate of Xiaohongshu reaches 26.5%, and the overall ROI is over 35% higher than that of traditional single-platform strategy, which verifies the validity and adaptability of the model in a multi-platform environment. Keywords: Short video marketing; Live e-commerce; User behavior modeling; Content strategy. 1. Introduction 1.1. Research Background In recent years, with the popularization of mobile Internet and the accelerated deployment of 5G technology, short video and live e-commerce have risen rapidly as emerging forms of content dissemination and merchandising, and have become an important force in promoting the development of digital economy. According to the 53rd statistical report released by China Internet Network Information Center (CNNIC), by the end of 2023, the scale of short video users in China had exceeded 1 billion, and the number of live e-commerce users exceeded 500 million, accounting for nearly half of the total number of Internet users. The platforms represented by Joyo, Shuttle, Taobao Live, and Xiaohongshu have built a new chain of "content-interaction-transaction", changing the traditional e-commerce shopping logic centered on search- driven. Content marketing has gradually become a key means of linking brands and users, and its value is not only reflected in the level of attracting traffic, but also lies in the deep impact on users' attention, consumption motivation and behavior path. In this context, the systematic analysis of the role of short video and live e-commerce content marketing strategy and user behavior is of great theoretical significance and practical value for improving marketing efficiency, optimizing resource allocation and extending the value of user life cycle. 1.2. Research Significance This study focuses on the analysis of content marketing strategy and user behavior in short video and live broadcasting e-commerce, which has important theoretical and practical double value. At the theoretical level, there is a lack of systematic modeling and analysis of the coupling relationship between content marketing and user behavior, especially in the multi- platform cross-environment, the mechanism of the influence of different content strategies on users' psychological cognition, behavioral paths and consumption transformation needs to be clarified. It is necessary to construct a content strategy model and user value assessment system to supplement the existing literature on the causal chain between content-behavior-commercial results. At the practical level, facing the challenges of e-commerce platforms' traffic dividend peaking and users' attention fragmentation, brands are in urgent need of accurate and replicable content operation methods to improve ROI and LTV, and empirical analyses of the differences between typical platform strategies and users' behavioral patterns provide feasible optimization paths and decision-making support for content design, platform operation, and marketing placement, helping enterprises to achieve the maximization of marketing efficiency under the content-driven approach. Maximization. 1.3. Literature Review Against the background of the rapid development of short video and live e-commerce, the study of content marketing and user behavior has gradually become the focus of academic attention. Zhang et al. (2025) [1] pointed out through the field survey of short video content in Yongxingfang, Xi'an, China, that short video marketing effectively stimulates users' interest and drives offline consumption behaviors by strengthening the digital footprint and immersive content experience, revealing a close relationship between content design and consumption conversion. Yuan et al. (2025) [2] analyzed the influence mechanism of community interaction on users' purchase intention in live e-commerce from the perspective of social learning theory, pointing out that the frequency of interaction, social trust and information sharing have a significant positive effect on users' decision-making, which provides theoretical support for the understanding of users' behavioral paths in the content environment. Guochao et al. (2025) [3] empirically analyze the key factors affecting consumers' purchase intention based on Chinese live e-commerce user data, and find that anchor professionalism, content authenticity and interaction density constitute important variables affecting consumers' 43 psychological expectations and purchase behavior, emphasizing the structural linkage between content and behavior. Guo et al. (2025) [4], on the other hand, take short travel videos as an entry point to construct a model of the content dissemination mechanism that affects users' travel behavioral intentions, verifying the mediating role of content presentation, information density and emotional guidance in the transformation of user behavior, and expanding the applicable fields and model boundaries of short-video content marketing. In addition, Jieli and Chunqing (2025) [5] studied the impact of content features on user engagement behavior in short video platforms, proposed that there is a nonlinear configuration effect between content structure complexity, emotional color and user value perception, and emphasized that different combinations of content strategies have a differentiated effect on user behavior activation. This study introduces fuzzy set qualitative comparative analysis (fsQCA) in methodology, which provides a new path for exploring the relationship between content features and user response. Current research has more systematically explored the association mechanism between content elements, user psychology and behavioral intentions in short video and live e-commerce, and constructed a relatively rich theoretical foundation. However, there are still three deficiencies in the existing literature: firstly, it focuses on a single platform or a single content dimension, and lacks cross-platform strategy differentiation research; secondly, the causal reasoning path between content strategy and user behavior has not yet been completely opened, and the behavioral modeling is still at the descriptive level; thirdly, there is a lack of a full chain of "attraction - transformation Third, there is a lack of integrated analysis framework for the whole chain of "attraction- conversion" and "repurchase". It is necessary to integrate content strategy modeling and behavioral mechanism analysis to deeply analyze the evolution logic and intervention path of content-driven marketing in a multi- platform context. 2. Analysis of the Current Situation of Short Video and Live E-Commerce Content Marketing 2.1. Analysis of Short Video Content Marketing Strategy Short video content marketing is algorithm-driven and interest matching as the core, through fragmented, high- frequency information delivery to achieve user attention capture and consumption transformation. Taking the Jittery platform as an example, its content distribution is based on the recommendation system constructed by user profiles, and through the "content vectorization + user behavioral similarity" model, it dynamically adjusts the weight of content exposure [6]. According to the "2024 Shake E- commerce Annual Data Report", the average conversion rate of short videos in the commodity category reaches 5.3%, of which the proportion of videos within 15 seconds is more than 72%, which indicates that short-duration, high-density content is more efficient for conversion. Content strategy usually combines contextual narrative, label reinforcement and visual guidance to optimize the user behavior path through the four-stage structure of "attraction-retention- guidance-action". Some brands have introduced A/B testing strategies to assess the impact of title, cover and script copy on CTR (click-through rate) and CVR (conversion rate) to maximize ROI in multiple rounds of placement. Figure 1 below shows a typical short video content marketing funnel model and the trend of key indicators. Figure 1. hort Video Marketing Funnel and CTR/CVR Comparison 2.2. Analysis of Live E-Commerce Content Marketing Strategy Live e-commerce content marketing strategy is centered on "strong interaction + scene restoration", driving users' instant purchase decisions through anchor persona, real-time interaction and limited-time promotions. According to the Ali Research Institute, the average conversion rate of Taobao live commodities in 2024 reached 8.2%, much higher than the 3.4% of traditional graphic e-commerce. Through carefully designed scripts and segmented rhythm, the anchor will divide the content structure into "context introduction - product display - interactive Q&A - limited time order", supplemented by pop-up interaction, lucky draws, coupons issued and other mechanisms, significantly increasing the length of user stay and the number of users. This significantly improves the length of stay and repurchase willingness of users [7]. The platform adopts a multi-dimensional scoring system (e.g., exposure E, interaction rate I, turnover T), and its comprehensive content score can be expressed as: S E I T  = + + (1) Where the weighting factor , ,   is dynamically adjusted based on the category and marketing objectives. At the content design level, head anchors emphasize visual performance and script control, while mid-waist accounts rely on category trust and accurate reach. 2.3. Case Study of Typical Platforms Jitterbug relies on strong algorithmic recommendation to realize the closed-loop marketing of "short content + strong transformation", with a click rate of 9.2%, suitable for instant decision-making consumption; Shutterbug focuses on the construction of trust in the private domain, emphasizes the long-term interactive relationship between anchors and fans, and the repurchase rate is as high as 27%; Taobao Live focuses on professional shopping guide, and drives the transaction through high-quality product selection and scenario-based scripting, with an average customer satisfaction rate of 27%. Taobao live focuses on professional shopping guide, through high-quality selection and scene script to drive the transaction, the average unit price is ahead of other platforms; Xiaohongshu to grass content as the core, emphasizing the user's real experience and community 44 dissemination, the average interaction rate of its notes reaches 5.6%, outstanding performance in brand awareness and high customer single product guidance. The strategies of the four platforms are different, applicable to different brand demands and user structures. 3. Content Marketing Strategy Model 3.1. Scenario-based Content Construction Scenario-based content construction is a key strategy for short video and live e-commerce content design by restoring real or simulated usage scenarios to enhance users' sense of immersion and purchase motivation. In short videos, product information is often embedded in specific scenarios such as life clips, workplace environments, outdoor applications, etc., to enhance the relevance of the content; in live broadcasts, product utility awareness is strengthened through scenery, role-playing and demonstrations of use [8]. Platforms usually adopt the scene matching degree model to measure the degree of fit between content and user interests, expressed as: 1 1 n i ii n ii w s M w = =  =   (2) Where is is the similarity between content labels and user interest labels, iw is the weight of each label, and M is the overall matching degree. The high matching degree scenario not only improves the click-through rate (CTR), but also increases the average page dwell time and user interaction frequency, which is a key path to realize the integration of precision marketing and content immersion experience. 3.2. KOL/KOC Cooperation Mechanism KOL/KOC cooperation mechanism builds an efficient content dissemination network through influence stratification and role differentiation: KOL (Key Opinion Leader) is suitable for building sound volume and guiding cognition as it has extensive exposure and brand endorsement; KOC (Key Consumer) is suitable for reaching market segments as it drives conversions with real usage experience and high trust. Brands often adopt the combination strategy of "head KOL momentum + waist KOC guidance + vegetarian diffusion" to realize the multi-stage influence path from awareness to conversion [9]. The platform evaluates the ROI of the cooperation based on the communication effect and optimizes the placement structure (see Figure 2). Figure 2. Livestream Content Structure and Conversion Flow 3.3. Real-time Interactive Content Design The design of real-time interactive content is a key link to improve the participation and conversion rate of live users, usually centered on the "frequency of interaction, interaction and feedback mechanism" to carry out a systematic layout. The design needs to be combined with the live rhythm, in the product explanation, peak conversion, near the end of the broadcast and other nodes embedded in the interactive module, such as pop-up questions, real-time polling, lucky draw, limited-time password red packets, etc., the formation of high-frequency contacts between the content and the user. Technically, relying on real-time data collection and user behavior monitoring, the interactive heat (H) is quantitatively managed, and the evaluation index can be set as: c g r I I I H t + + = (3) Where Ic, Ig, Ir is the number of comments, likes and retweets respectively, and t is the length of the time window. Highly interactive content is usually synchronized with rolling incentive feedback (e.g., instant winners, exclusive coupon push) to enhance users' motivation to participate and willingness to stay. 3.4. Personalized Content-Driven Strategy Personalized content-driven strategy relies on the construction of user profiles and behavioral data analysis to achieve accurate matching and targeted delivery of content. The platform builds a multi-dimensional image system by collecting user browsing records, length of stay, clicking behavior, historical purchase data and other information, and dynamically generates a content recommendation pool by combining collaborative filtering and deep learning recommendation models [10]. Specific strategies include video script personalized adjustment, title keyword optimization, content sorting customization, etc., to meet the differentiated user preferences. In the live broadcast scenario, the system can automatically push the corresponding category live broadcast according to the user's interest label, and dynamically adjust the pop-up window content of the live broadcast and the form of the offer to achieve real-time adaptation of the content and the crowd, so as to improve the conversion efficiency and user satisfaction. 4. User Behavior Modeling Analysis 4.1. Behavioral Feature Extraction Commonly used indicators include click-through rate (CTR), browsing depth, dwell time, forwarding rate, add-to- buy rate, bounce rate, order conversion rate, etc. The variables reflect the intensity of users' perception and behavioral willingness to different content forms. The platform records every slide, stay, click, and interaction of the user in the video or live broadcast in a fine detail through the event log collection system, and combines the timestamps with the content features for sequence coding. Part of the research introduces sequence pattern mining and clustering algorithms to identify high-frequency behavioral paths, realizing the behavioral chain tracking from "browsing→interest→interaction→conversion". The structured extraction and normalization of behavioral features can be used for subsequent training of user segmentation, value assessment and personalized content recommendation 45 models, providing data support for precise strategy implementation. 4.2. Behavioral Differences Between Platforms Different platforms show obvious differences in the distribution of user behavioral characteristics due to differences in content ecology, recommendation mechanism and user positioning. In Jitterbug, users are mainly interest- driven, consumption decisions are more impulsive, frequent content skipping, high CTR but short stay time; fast hand users are sticky, pay more attention to the anchor relationship and word-of-mouth accumulation, the conversion path is longer but the repurchase rate is high; Taobao Live focuses on the professional guide to the purchase of commodities, the user's goal is clear, the conversion efficiency is high, but the frequency of interaction is low; Xiaohongshu focuses on the seeding of grass content, the user prefers in-depth reading and collection, the behavioral path is expressed as ""," and the behavior is characterized by """. Xiaohongshu focuses on grass content, users are more inclined to read in depth and collect, the behavioral path is "delayed conversion type". Table 1 below compares the performance differences of the four platforms in terms of core user behavior indicators. Table 1. User Behavior Metrics Across Platforms Platform CTR (%) Avg. Stay Time (s) Add-to- Cart Rate (%) Repurchase Rate (%) Douyin 9.2 18.4 5.6 14.7 Kuaishou 7.6 25.9 6.9 27.3 Taobao Live 6.3 21.1 8.2 18.5 Xiaohongshu 4.5 32.6 3.1 11.4 Table 1 shows that there are significant differences in user behavior indicators across platforms. Jitterbug tops the list with a click rate of 9.2%, reflecting the efficient capture of users' attention by its strong interest recommendation mechanism, but the average length of stay is only 18.4 seconds, reflecting the fast pace of users' decision-making. Snapdragon, on the other hand, excelled in its repurchase rate, reaching 27.3%, indicating stronger user stickiness driven by its private domain relationship. Taobao Live has an 8.2% repurchase rate, which is higher than other platforms, showing that its professional shopping guide content has a significant effect on conversion. Xiaohongshu users stayed for an average of 32.6 seconds, indicating that its in-depth content is more likely to lead to long-term cognitive accumulation and delayed conversion (see Figure 3). Figure 3. User Behavior Metrics Across Platforms 4.3. User Value Modeling User value modeling is an important means to measure the contribution of different users to the platform's business, which is usually assessed from the dimensions of lifecycle value, purchase frequency, customer unit price and behavioral activity. In the short video and live e-commerce environment, user behavior paths are highly fragmented, and it is necessary to dynamically identify and stratify the potential value of users by combining their successive actions, such as browsing, clicking, purchasing, ordering, and repurchasing. High-value users often have high activity, stable consumption behavior and strong brand stickiness, which is suitable for directional push of high-conversion rate content; low and medium-value users need to enhance their activity and conversion potential through content optimization and interactive guidance. The platform often divides users into four categories: "attraction type", "potential type", "conversion type" and "loyalty type", and The users are categorized into four types: "attracting", "potential", "converting" and "loyal", and are matched with different content strategies and marketing channels. Through the systematic modeling of user value, we can realize the refined allocation of resources and maximize marketing efficiency, effectively improving the overall ROI of the platform and user life cycle revenue. 4.4. Causal Relationship Between Strategy and Behavior In the short video and live e-commerce environment, the relationship between content strategy and user behavior is not a simple correlation, but has a clear causal chain. Different types of content strategies, such as contextual narrative, emotional rendering, interactive guidance and time-limited promotions, stimulate users' interest, emotional resonance, willingness to participate and impulse to buy, thus triggering continuous behavioral responses. For example, the "before and after" content in short videos can significantly increase users' click-through rate and length of stay, while the "limited- time discount + pop-up lottery" combination in live broadcasts can effectively promote users' ordering behavior 46 in a short period of time. It is found that when the content strategy is highly matched with the user's current interests and needs, the conversion rate of user behavior will increase significantly, and vice versa may lead to an increase in the bounce rate. Therefore, understanding the causal mechanism of strategy design on user behavioral changes helps the platform to realize the whole chain optimization from content creation to behavioral intervention, and improve the accuracy and predictability of marketing operations. 5. Case Validation and Path Optimization 5.1. Case Setting This study selects the integrated marketing campaign of a national beauty brand to promote its new lip glaze on the three platforms of Jittery, Taobao Live and Xiaohongshu as a case study, focusing on the whole process of "attraction- conversion-repurchase" content strategy and user behavior response. The specific operation process is as follows: put a 15-second short video of the product on Jittery Voice, set interest tags and enable Qianchuan ads to divert traffic to Taobao Live; the head anchor explains the product and promotes conversion with limited-time discounts, and at the same time publishes seeding notes on Xiaohongshu to strengthen the product's word-of-mouth and the user's cognitive precipitation. The data collection covers the video clicks, the number of live viewers, the conversion rate of the order, the Little Red Book collection and comments and other dimensions, forming a content-behavior closed-loop data chain, and combined with the user behavior logs, to establish a causal path analysis model for verifying the effect of the content strategy on the various behavioral nodes. 5.2. "Attraction+Conversion+Repurchase" Model The "attraction + conversion + repurchase" model is a three-stage content marketing system based on user behavior paths, aiming to realize the optimization of the whole process of user operation on short video and live e-commerce platforms. In the "attraction" stage, the platform realizes user interest stimulation and traffic aggregation through short video content placement, and often attracts users' clicks and concerns with high-exposure and emotionally contagious videos, guiding them to the live broadcasting room or the brand's homepage; in the "conversion" stage, the platform uses live broadcasting for real-time explanation and limited- time promotions. In the "conversion" stage, through the live real-time explanation, limited-time promotions and interactive incentives to strengthen the user's willingness to buy, combined with the product demonstration and benefit points conveyed to shorten the user's decision-making chain and improve the conversion rate of the order; in the "re- purchase" stage, relying on the community platforms, such as Xiaohongshu and other community grass-roots content and the brand's secondary content to reach, through the user's comments, true experience In the "repurchase" stage, it relies on community platforms such as Xiaohongshu and the brand's secondary content reach, through user comments, real experience and repeated exposure of the content to build a cognitive closed loop and enhance the user's long-term trust in the brand and repurchase frequency. This model realizes the flow and precipitation of users among multiple platforms through content density control and behavioral rhythm management, taking into account short-term sales conversion and long-term user value, with high adaptability and replicability, and has become a typical path structure for multi-platform integrated marketing. 5.3. Cross-Platform Collaboration Path Cross-platform synergistic path is a key strategy to enhance the efficiency of short video and live e-commerce content marketing. By linking user behavior and content strategy, it realizes the efficient synergy of "attraction-conversion-re- purchase" among multiple platforms. Specific path design follows the principle of "complementary content styles, misplaced user roles, and complementary functional advantages": in the Jitterbug stage, algorithmic recommendations drive high-exposure short videos to quickly achieve interest reach and initial attraction; Taobao live streaming takes on the task of conversion, through professional anchor explanation and limited-time promotions to strengthen users' purchasing behavior; Xiaohongshu is Xiaohongshu focuses on user seeding and cognitive precipitation, and promotes long-term repurchase and word- of-mouth communication through real experience sharing and high-trust social networks. At the technical level, the platforms achieve user identification and content adaptation through user account binding, advertising pixel embedding and behavioral label synchronization; the brand side ensures the consistency of the content style and objectives of each platform through a unified content strategy and sub-platform operation division of labor, avoiding duplicate exposure and waste of resources. In addition, the synergistic path also needs to work with the data attribution mechanism to identify the weight of different platforms in the final conversion, so as to dynamically optimize the proportion and rhythm of content placement and maximize the effectiveness of cross-platform synergy. This mechanism has been validated as an effective strategy to improve ROI and LTV by many head brands. 5.4. Effect Evaluation In order to comprehensively assess the actual effectiveness of the cross-platform content marketing path of "Jittery voice attraction + Taobao conversion + Xiaohongshu planting", this paper selects the data of a national beauty brand's integrated promotional activities on the three platforms for comparative analysis, covering the key indicators of attraction efficiency, conversion effect, re-purchase behavior and comprehensive ROI. Data sources include platform advertising background, live transaction data, user behavior logs and content interaction records. The evaluation period is set at 30 consecutive days after the campaign is launched, and the statistical indicators are shown in Table 2 below. 47 Table 2. Performance Metrics of Cross-Platform Content Marketing Campaign Platform Exposure (10k) CTR (%) Avg. View Time (s) Conversion Rate (%) Avg. Order Value (¥) Repurchase Rate (%) Douyin 865 8.9 15.6 2.7 — — Taobao Live 62.4 6.1 24.3 9.6 138.5 18.2 Xiaohongshu 27.8 4.7 31.9 3.4 146.2 26.5 As can be seen in Table 2, Jieyin as a lead generation platform realized a total of 8.65 million exposures with a click rate of 8.9%, accumulating a large amount of initial traffic for the conversion segment. Taobao live streaming completed the main sales task with a click rate of 6.1% and a conversion rate of 9.6%, with an average order amount of RMB 138.5, indicating that professional content output significantly enhances users' purchase intention. Although the click rate of Xiaohongshu is only 4.7%, its average stay time reaches 31.9 seconds, and the repurchase rate is as high as 26.5%, playing a key role in the long-term operation of users. 6. Conclusion The evolution of short video and live e-commerce content marketing strategy has moved from single-platform operation to a new stage of cross-platform synergy, the essence of which lies in the deep integration between content, behavior and technology. This paper builds a theoretical model based on the path of "attraction - conversion - repurchase", and systematically analyzes the guiding effect of scenario-based content construction, KOL/KOC synergistic mechanism, real-time interactive design and personalized content delivery on user behavior, and through the case studies of Jittery, Taobao, and Xiaohongshu, the content marketing strategy has evolved from single-platform operation to cross-platform synergy. We also verify the causal mechanism and structural difference between multi-platform content strategy and behavioral response through the empirical evidence of the cases of Shakeyin, Taobao, and Xiaohongshu. The empirical results show that efficient content combination strategy and accurate user stratification operation can not only significantly improve ROI and conversion rate, but also extend the value of user life cycle and realize the optimal allocation of marketing resources. Future research can further introduce multi-dimensional contextual variables and real- time behavioral feedback mechanisms, combine AIGC content generation and multimodal identification technologies to expand the intelligent adaptation capabilities of content design, and establish more mature mechanisms in data attribution, security compliance and algorithm transparency to support a more sustainable and explainable content-driven business model. References [1] Zhang J ,Chan S C . Surviving from the craze: Short video marketing on digital footprint in Yongxing Square in Xi’an [J]. Journal of Vacation Marketing, 2025, 31 (3): 562-577. [2] Yuan Y ,Bingfeng B ,Sihua X . The role of social community in influencing purchase intention in live-streaming E- commerce: a social learning theory perspective [J]. Marketing Intelligence & Planning, 2025, 43 (5): 934-951. [3] Guochao P ,Fei X ,Xiaodan Z . Key Factors Affecting Consumers' Purchase Intention in Live Streaming E- Commerce: An Empirical Study in China [J]. Journal of Global Information Management (JGIM), 2025, 33 (1): 1-20. [4] Guo X ,Liu B ,He J , et al. The impact mechanism of tourism short video content marketing on users’ travel behavioral intention [J]. Humanities and Social Sciences Communications, 2025, 12 (1): 494-494. [5] Jieli Z ,Chunqing L . The influence and configuration effect of content characteristics on customer input in the context of short video platforms [J]. Journal of Research in Interactive Marketing, 2025, 19 (3): 482-497. [6] Saier H . The Influence of Tourism Public Relations Strategies on Urban Brand Reputation via the Chinese Douyin Short Video Platform [J]. Studies in Media and Communication, 2025, 13 (2): 159-174. [7] Yuan W . A Review of Content Marketing’s Influence on Consumers’ Purchase Intention in Live-streaming E-commerce [J]. Proceedings of Business and Economic Studies, 2025, 8 (1): 55-59. [8] Zhang Z ,Chen Z ,Wan M , et al. Choice of product quality in supply chain of live-streaming e-commerce under different power structures [J]. Australian Journal of Management, 2025, 50 (1): 266-286. [9] He D ,Assoc ,Hashim B F D P . Quantitative Analysis of Live E-commerce's Influence on Consumer Purchase Intentions [J]. Academic Journal of Management and Social Sciences, 2024, 9 (2): 239-245. [10] Sumarlam S ,Djatmika D ,Widiana Y , et al. Persuasive Speech Act Strategies of Online Fashion Sellers in Live E-Commerce: A Cyberpragmatics Approach [J]. Theory and Practice in Language Studies, 2024, 14 (11): 3385-3393. [11] Wu B Y. Market research and product planning in e-commerce projects: A systematic analysis of strategies and methods,.Academic Joual of Business & Management, 2025, 7(3): [07].