Descriptive Analysis of Impulsive Purchase Intention on Live-Streaming Commerce in Indonesia

. This research aimed to discover the behavior of impulsive purchase intention on live streaming e-commerce in Indonesia based on Interaction quality and IT Affordance. This research used a quantitative approach to descriptive analysis. The samples used were 532 respondents. The researchers distributed the questionnaire to social media and online messaging applications. Based on the demographic profile of respondents, the researchers discovered that the respondents are highly educated, work as professional employees or college students, and allocate more than once per week to shop on e-commerce platforms. Based on the live streaming e-commerce behavior of the respondents, the respondents are active on live streaming, just watching less than four hours per week. The respondents require good interaction quality during live streaming, such as quick responses from anchors, in-depth product information from anchors, best product solutions from anchors, and an excellent live streaming atmosphere from anchors. The respondents also need good IT Affordance during live streaming, and e-commerce management must provide high-level visibility features, good communication features for all users, and features that help the anchor respond to customer product requests.


INTRODUCTION
In this increasingly advanced technological era, people can take advantage of online shopping systems through digital platforms such as e-commerce.E-commerce has implemented a new shopping method, namely the live streaming system.This new mode of e-commerce uses live streaming to engage customers on e-commerce platforms, whereby broadcasters utilize new media as a direct source for online sales [1].The COVID-19 pandemic boosted live-streaming shopping.China sold $200 billion in live streaming during the 2020 pandemic, while the US sold $11 billion [2].In June 2022, Shopee had 83.4% of Indonesian shoppers, followed by Tokopedia at 30.4% and Lazada at 20.5% [3].78% of Indonesian shoppers are aware of alternatives to live-streaming.71% of consumers used it, and 56% bought things via live broadcasting during the pandemic [4].On 11.11, 2021, Chinese megainfluencer Austin Li generated $1.8 billion in sales from live streaming on Alibaba e-commerce in 12 hours [5].This proves that people are starting to change their shopping activity schemes in e-commerce from traditional to modern with live streaming features.Adopting a live streaming system will increase sales from e-commerce itself and even improve the trading economy of the country.Therefore, exploring the factors influencing consumers' impulsive purchase intentions in shopping on live streaming is necessary.
However, to make live streaming commerce the main contribution, several challenges must be considered in its implementation.From the buyer's perspective, the challenge is that the quality of goods in live streaming is uneven.Consumer rights face various obstacles, such as unclear live titles and communication difficulties [6].All these problems can reduce consumer confidence in e-commerce platforms and sellers, which will ultimately affect the economic development of live shopping.Then from the seller's perspective based on Beck,2021 [7], the first challenge faced is consumer unpreparedness, where some buyers are still accustomed to searching for products and comparing product reviews independently.Moreover, according to them, watching live streaming is time-consuming and less efficient.Business actors must convince buyers about the benefits of shopping on live streaming.The second challenge is customer retention, where businesses must innovate new ways to retain and bring consumers back to shop on live streaming.The third challenge is the high costs, where products are usually sold at significant discounts on live streaming, which is an expensive form of investment.
Based on existing research, the factors influencing impulsive purchase intentions can be divided into consumer, product, and environmental factors.Consumer factors are studied from the perspective of emotions [8] and personality traits [9].Product factors are studied from the perspective of product types [10].Environmental factors are studied from the perspective of the live atmosphere [6], online reviews [11], and IT affordability [12].Several of these studies provide references for this research to explore the factors influencing impulse buying intentions in shopping on live streaming.
As researchers conducted the literature review, there are limited study that develop comprehensive factors of live streaming shopping from interaction quality and IT affordance perspective in South Asia, especially Indonesia context.Previous research in Indonesia states that the IT affordance variable is significant in live streaming [12,23].As for the interaction quality variable, no one has studied it in Indonesia, so this study will adopt the S-O-R (stimulus-organism-response) framework to examine consumer behavior in live streaming.This research introduces the situation factors as a stimulus that affects the customer's affective and cognitive perceptions.We adopt responsiveness, professionalism, informativeness, and personalization to arousal emotion (affective reaction) [16].We assume visibility, meta voicing, and guidance shopping to Immersion (cognitive response) [12,23].
Further, we adopt impulsive purchase intention to investigate the final impulsive purchase intention.The addition of these variables is an element of novelty in research in Indonesia compared to previous studies.With the limitations of prior research, the main objective of this research is to discover the behavior of impulsive purchase intention on live streaming e-commerce in Indonesia based on Interaction quality and IT Affordance.

METHODS
This research uses a quantitative approach to descriptive analysis.The population in this research is Live-Streaming users who have watching live-streaming in ecommerce Indonesia.The samples used are 532 respondents.The sample size determination method is ten times the number of indicators [26].Then, convenience sampling is also used as the sampling technique in this research.The researchers create an electronic questionnaire using Google Forms and distributed it through social media (Instagram, Twitter, and Facebook) and Online Messaging Applications (Line, WhatsApp, and Telegram) to Circles of friends, family, Communities, and some relatives.The authors used an application called Stats Tools Package through Microsoft Excel 2016 as a software for analysing the questionaire data.
There are ten variables in this research.All indicators were adapted from the previous research and measured using likert scale.Three indicators of responsiveness adapted from Jiang et al. [27], Three indicators of profesionalism adapted from Brady and Conin [28], Three indicators of personalization adapted from Winsted [29], Three indicators of informativeness adapted from Mathwick et al. [30], Four indicators of visibility adapted from Saffanah et al [23], Four indicators of metavoicing adapted from Saffanah et al [23], Three indicators of guidance shopping adapted from Saffanah et al [23], Three indicators of arousal emotion adapted from Koo and Ju [31], Five indicators of immersion adapted from Yim et al [32], and Four indicators of impulsive purchase intention adapted from Lee and Chen [33].
The data was analyzed and presented into two perspectives.The first perspective, the researchers presented the demographic profile of respondents.The researchers used frequency and percentage to show the data analysis results of demographic descriptive analysis.The second perspective, the researchers presented the questionnaire items for explaining the performance of live streaming online shopper view.The researchers used mean, median, and standard deviation to show the data analysis results of variable descriptive analysis.

RESULT AND DISCUSSION
The result provides an interpretation of demographic profile of respondents, live-streaming e-commerce respondents, and descriptive analysis of conceptual variables.After the descriptive analysis, the theoretical construct is measured statistically.The descriptive information includes the demographic profile of respondents, such as ages, occupation, Frequency watching live streaming (1 week), and Frequency shopping at e-commerce (1 week).Table 3 describes variables and their indicators are performed using mean, median, and standard deviation statistics.The mean of responsiveness variables is 4,803.It can be said that live-streaming users in Indonesia feel that the anchor is willing and patient to respond to customer questions in a relevant manner.Then, the mean of professionalism variables is 5,088.It implies that live-streaming users in Indonesia feel that anchors have communication skills and professional product knowledge when explaining products.The mean of informativeness is 4,866.It means that livestreaming users in Indonesia get valuable information and deepen their understanding of the product when the anchor provides product-related information in livestreaming e-commerce.Meanwhile, the mean of Personalization variables is 4,881.It implies that the high service anchors can help customers get goods according to their needs through product suggestions in live-streaming e-commerce.For arousal emotion, the mean is 4,732, which means that the high arousal emotion can make customers feel excited when browsing and buying products in live-streaming ecommerce.Table 3 describes variables and their indicators are performed using mean, median, and standard deviation statistics.The mean of visibility variables is 5,118.It can be said that the high visibility in live-streaming ecommerce in Indonesia can make product demos visible to customers and make customers feel what the product looks like in the real world.Then, the mean of metavoicing variables is 4,931.It implies that ecommerce live streaming in Indonesia allows customers to provide feedback and interact with anchors through the comment's column feature.The mean of guidance shopping is 4,836.It means that live-streaming users in Indonesia feel that the anchor in live-streaming can provide solutions in the form of alternative products and help choose products according to their needs.For the immersion variables, the mean is 4,515, meaning that live-streaming users in Indonesia find live-streaming platforms interesting.Customers feel involved when they watch live streaming in e-commerce.Based on descriptive analysis of variables, the implications for the anchors, the sellers, and ecommerce management must focus on the immersion aspects of live-streaming shopping.There are several proposed suggestions for e-commerce management and the anchors to increase the users impulsive purchase behaviour, arousal emotion, and immersion.First, ecommerce management must catch users' attention by sending notifications on applications and social media regarding products on live streaming.Then, distribute it to users who have never watched a live stream and map it based on the user's search history behavior so that the live streaming content displayed will match the results of their previous product search.Second, the anchor must actively interact with the customer and show enthusiasm and energy during the live stream to increase the customer's emotional arousal [16,34].Then, ecommerce management can develop music and sound effects to create an emotional connection with customers, such as upbeat music that will create an excited and surprised atmosphere for customers in live streaming.Thrid, the anchor and e-commerce management can set up well-designed virtual stores within live streaming.A well-designed virtual store can help customers feel they are shopping in a physical store.Stores must be easily accessible, and products must be displayed regularly and attractively.
Furthermore, several proposed suggestions exist for e-commerce management and the anchors to increase interaction quality during live streaming.First, anchors must respond quickly to questions posed by customers [16].The seller must set up moderators to assist anchors in monitoring comments and questions.E-commerce management can develop chatbot features to help anchors answer frequently asked questions so the anchors can focus on more complex questions [35].Second, the anchor must improve presentation skills, such as body language and facial expressions, to create a good atmosphere in the live-streaming room [16,36].The anchor can also create interactive content, such as quizzes and games related to the products they are showcasing.E-commerce management can develop virtual and augmented reality technologies in live streaming so the anchor can use these technologies to present products in an immersive and attractive way [32].Third, the anchor must provide accurate and detailed information to customers [16,37].E-commerce management must provide detailed product information accessible to customers in the live-streaming room.Last, the anchor must interact with customers by finding customer needs and providing product information that fits customer needs [16].The anchor can also follow up with customers after the live-streaming ends, such as conducting personal chats with customers to provide the desired solution [38].
Finally, there are several proposed suggestions for ecommerce management, the anchors, and the seller to increase IT Affordance during live streaming.First, the sellers should invest in high-quality cameras and lighting equipment to ensure clear and detailed images are captured during the live stream.This can enhance the viewing experience for customers and increase customers perceived immersion [12,23].Next, ecommerce management can provide high visibility with pre-recorded product video features that display the product in detail.These videos can be played during live streams to give customers a better view of the product.Second, E-commerce management must develop metavoicing attributes such as customer interaction features via chat features [12,39].Then, e-commerce management can develop social media sharing features and Q&A sessions during or after a live shopping event where customers can ask questions about products and get answers from the anchor or other customers.Last, the sellers can identify the most common customization requests for customer products and create preconfigured options for customers during live streaming.These options can be based on product color, size, or other attributes.Then, e-commerce management can develop product configuration features during live streaming to enable customers to select the desired product options in real-time and unlimited.It should be easy to use and provide real-time updates on product images or prices based on the options selected.

Conclusion
Based on the demographics of respondents, the researchers discover that the respondents are highly educated, work as professional employees or college students, and allocate more than once per week to shop on e-commerce platforms.Based on the live streaming ecommerce behavior of the respondents, the respondents are active on live streaming, just watching less than four hours per week.Moreover, for descriptive analysis of the research variable, the respondents require good interaction quality during live streaming, such as quick responses from anchors, in-depth product information from anchors, best product solutions from anchors, and an excellent live streaming atmosphere from anchors.The respondents also need good IT Affordance during live streaming, and e-commerce management must provide high-level visibility features, good communication features for all users, and features that help the anchor respond to customer product requests.The respondents felt excited and surprised when browsing and buying products in live streaming ecommerce.When joining the live streaming room, the respondents found the live-streaming platform attractive and they felt involved while watching live streaming.This will strengthen the respondents' feelings of having impulse purchase intentions when shopping on ecommerce live streaming.
This research only uses descriptive analysis to understand the impulsive purchase intention from the

Table 2
describes variables, and their indicators are performed using mean, median and standard deviation statistics.The mean of impulsive purchase intention to buy products recommended by anchors via livestreaming e commerce in Indonesia.

Table 2 .
Mean, Median, and Standard Deviation Statistics of Variables and Indicators Impulsive Purchase Intention

Table 3 .
Mean, Median, and Standard Deviation Statistics of Variables and Indicators Interaction Quality

Table 4 .
Mean, Median, and Standard Deviation Statistics of Variables and Indicators IT Affordance