The Effect of Conversion Rate on Product Sales from the Perspective

of AISAS Model: An Empirical Study based on the Data of the

Amazon Merchants

Jing Li

, Xue-Liang Wang

and Ye Wang

1,*

School of Management, Guangzhou Xinhua University, Guangzhou, China

Guangzhou Yin-Chen-Xing Trading Co., Ltd., Guangzhou, China

Corresponding author

Keywords: Cross-Border E-Commerce, Amazon Platform, Conversion Rate.

Abstract: With the rapid development of cross-border e-commerce, Amazon platform, as the largest and most mature

cross-border B2C platform, has attracted a large number of merchants. In the face of fierce competition, how

to increase product sales is the most difficult problem for every Amazon merchant. Based on the AISAS

model of consumption to the amazon online store goods in a company as the research object, for the various

stages of conversion selected independent variables, the empirical analysis, and hypothesis testing, explore

the amazon online store goods at different stages of conversion rate and the relationship between sales,

according to the different conversion rate's impact on sales of amazon operating process are proposed. This

study collected the conversion rates of products in different stages, and draws the following conclusions: the

rate of page clicks, the rate of buy box and conversion rate of products have a significant positive effect on

sales volume.

1 INTRODUCTION

The conversion rate of goods at various stages can

reflect the level of the seller's store in the operation

process. The research on the conversion rate of online

goods plays an important role in the operation of e-

commerce enterprises. Commodity data can be used

to develop new products, and it can also predict the

trend and prospect of the market in the future. Based

on the first-hand data of an Amazon merchant, we

study the effect of commodity conversion rate on

sales volume at various stages in the operation

process. By studying the relationship between the

conversion rate and sales volume of amazon online

products, we can help cross border e-commerce

(CBEC) enterprises in different stages to optimize

and improve their operation ability. According to the

different conversion rates, the small and medium

CBEC companies can develop appropriate strategies

to promote products with low human and time costs,

improve product conversion rate and increase order

quantity.

2 LITERATURE REVIEW

2.1 E-consumer Purchasing Behavior

Consumer behavior is the process of consumers’

experience or ideas when choosing and using

products or services, which has an effect on the

consumers and society (Sabine 2012). Consumer

behavior includes pre-purchase activities and post-

purchase consumption, evaluation and processing

activities. Under the traditional conditions,

consumers have the characteristics as the following:

infinite demand, inducibility, variability and multi-

level. Because the Internet has greatly changed the

way of consumer behavior, its characteristics are also

changed as the following: personalization of

consumer products, convenience of consumption

pattern, rationalization of consumer behavior,

interactivity and initiative of consumers. These

changes have effect on the purchasing behavior and

decision-making of consumers.

In the background of the Internet, consumers

online purchasing behavior, characteristics and habits

of E-consumers have become the main research

556

Li, J., Wang, X. and Wang, Y.

The Effect of Conversion Rate on Product Sales from the Perspective of AISAS Model: An Empirical Study based on the Data of the Amazon Merchants.

DOI: 10.5220/0011191300003440

In Proceedings of the International Conference on Big Data Economy and Digital Management (BDEDM 2022), pages 556-562

ISBN: 978-989-758-593-7

objects of many scholars. For example, Zhi-Cheng Li

builds a new model based on the theory of planned

action. He studied from the perspective of the

characteristics of consumer behavior, carried out the

research on the influencing factors of online

consumers’ purchasing behavior, which had certain

guiding function to the practice of B2C (Li 2002). In

order to study the influencing factors of consumer

behavior in the context of big data, Guan-Ting Zhu

established the theoretical model of C2C, tested the

hypothesis, drew conclusions and gave constructive

suggestions (Zhu 2015). Na Zhou et al. used the R-

type system clustering method to investigate and

empirical the influencing factors of online

consumers’ purchasing decisions and came to the

conclusion that the eight factors, such as word-of-

mouth, brand and sales volume, were the key

influencing factors in the process of consumers ’

purchasing decisions(Zhou 2017).Based on SOR

model (stimulus-organic-response), Zan Mo et al.

studied whether online product evaluation would

have an effect on consumer behavior from the

perspective of consumer learning, and the final study

showed that positive evaluation had a positive

effect(Mo 2015).

2.2 AISAS Model

AISAS model is a brand-new consumer behavior

analysis model proposed by Dentsu Company in

2005, which aims at the changes of consumer life

style in the era of Internet and wireless application. It

emphasizes the relationship between all aspects of

interrelated and mutually influence each other

constraints with user experience. AISAS model

consists of five parts: Attention -- Interest -- Search -

- Action -- Share, that is, consumers gradually change

from the original passive receivers of information to

the receivers of independent information collection.

AISAS model emphasizes that online consumers will

spontaneously use third-party shopping platforms,

community apps and search engines to search the

product information that they are interested in, and

consumers will take the initiative to share user

experience and product quality information with

others after making purchases.

Third-party platforms or websites can help users

better understand products or change or influence

their purchase decisions. They can also promote the

interaction and communication between people.

Enterprises can use consumers' online behaviors to

quantify and digitize data, study, analyze and

interpret the data, and the conclusions drawn from the

data can help enterprises or operators to develop

suitable and feasible marketing strategies.

In the current research on commodity conversion

rate, most scholars mainly focus on the conversion

rate of a specific module in the operation of online

stores, and seldom study the conversion rate of each

stage in the operation process. For example, Bao-Wu

Bian et al. studied the commodity attraction factors of

the conversion rate of enterprise e-commerce

websites, and found that the factors affecting the

conversion rate of enterprise e-commerce websites

include website brand, commodity attraction,

customer service, customer behavior, user

experience, flow quality and other factors (Bian,

2019). Jing Jiang and Zhi-Yong Yang found that

attention and intention had a significant effect on

sales conversion. They suggested upgrading

communication methods according to the experience

model, and in-depth communication could promote

sales conversion (Jing, 2013). Tan Kai Yee discussed

the conversion rate of Amazon products based on the

AISAS consumer model, and made a regression

estimate of products and sales volume to help

improving the application value and assisting

enterprises engaged in cross-border e-commerce to

solve the problems of their products (Tan, 2018).

Therefore, based on the first-hand data of the

background of a company's Amazon website, this

paper studies the effect of product conversion rate on

sales volume in various stages of operation.

3 THEORETICAL MODELS AND

HYPOTHESES

3.1 Theoretical Model

Conversion rate means the percentage of users who

do positive behavior to the webpage versus all users.

The behavior of users in web can be quantified, such

as browsing, click, buy, evaluation, etc. The

conversion rate outwardly is a number, but as the

growth of the Internet websites and platforms, we can

find conversion rate reflects the Internet websites or

platforms a lot of problems. Conversion rate is

particularly important in the transactions of Internet

third-party platforms. To some extent, it is the

cornerstone of the growth of Internet platforms. High

conversion rate can bring greater returns to

enterprises. AISAS model, which was reconstructed

based on e-market characteristics in the Internet era,

is composed of five stages. That is Attention- Interest

- Search – Action-Share. Consumers search for

The Effect of Conversion Rate on Product Sales from the Perspective of AISAS Model: An Empirical Study based on the Data of the

Amazon Merchants

557

commodities after noticing them and becoming

interested in them, and share the information after

purchasing them. In this process, consumers cannot

do anything without the applications of the Internet.

Based on the AISAS model and the purchasing

behavior of online consumers, this paper builds a

theoretical model to reflect the changing process of

conversion rate at different stages in the AISAS

model, as shown in Figure 1.

Figure 1: Conversion Rate & AISAS Model.

3.2 Data Sources

The main source of data collection in this paper is the

background data of a company's Amazon online

store. In this paper, we selected the data collections

which were the conversion rates of 16 apparel

products in different stages from November 1, 2019

to December 31, 2019.

3.3 Research Hypothesis

Based on previous studies, this paper explains the

seven variables in the research model operationally,

so as to facilitate the understanding of consumers'

activities in the process of purchasing behavior, as

shown in Table 1.

Table 1: Variable Explanation.

Variables Explanation

Click rate

Refers to the percentage of visitors on a web page who make positive actions for the

enter

rise

Browse conversion rate

Refers to the effective conversion rate of an enterprise's products from exposure to

viewing.

Purchase conversion rate Refers to the conversion rate of consumers from effective browsin

urchase.

Buy button

win rate

Buy Box in Amazon refers to the golden shopping cart, which is the habitual place

for buyers to purchase on Amazon. The purchase win rate refers to the ratio of the

shopping cart in hand. If the table 100% indicates that the shopping cart is absolutely

in hand, Amazon buyers will give priority to purchase your products in the first time.

Favorable conversion rate

Refers to the percentage of positive reviews given to a product after it has been

urchased by customers.

Refund Conversion Rate

Refers to an effective proportion of products that customers choose to return due to

various reasons after purchase. Although the refund rate cannot be directly reflected

to consumers, enterprises can make timely countermeasures based on the feedback

of consumers.

Sales volume The number of

roducts actuall

sold in a

iven

eriod of time.

This paper defines sales volume as dependent

variable and other variables as independent variables,

discusses the relationship between conversion rate

and sales volume, studies its influence, analyzes the

influence of conversion rate on sales volume in

different stages, and puts forward constructive

marketing suggestions for enterprises based on the

influence of conversion rate on sales volume. The

research hypotheses of this paper are listed as follow:

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

558

Hypothesis 1: the webpage click conversion rate

at the stage of consumers' interest has a significant

effect on product sales.

Hypothesis 2: the browse conversion rate in the

consumer search stage has a significant effect on

product sales.

Hypothesis 3: The purchase conversion rate

during the consumer action stage has a significant

effect on the sales volume.

Hypothesis 4: the conversion rate of favorable

comments in the consumer sharing stage has a

significant effect on product sales.

Hypothesis 5: The conversion rate of refund in

consumer sharing stage has a significant effect on

product sales.

Hypothesis 6: The winning rate of buy button in

consumer action stage has a significant effect on

product sales.

4 RESULTS AND HYPOTHESIS

TESTING

4.1 Correlation Analysis

Table 2: Correlation Analysis

sales

Click on the

page

Conversion rate

Page views

Conversion

rate

The buy

button

Win rate

buy

Conversion

rate

Favorable

comment

Conversion

rate

A refund

Conversion

rate

sales

Pearson

correlation

1 736.

.736

.023 .391 .175 -. 278

Significance

(bilateral)

001. .001 .932 .134 .518 .297

16 16 16 16 16 16 16

Page click

conversion

rate

Pearson

correlation

.736

1 1.000

-. 474 - .057 -. 208 -. 214

Significance

(bilateral)

.001 .000 .064 .835 .440 .425

16 16 16 16 16 16 16

Page view

conversion

rate

Pearson

correlation

.736

1.000

1 - .479 - .052 -. 215 -. 206

Significance

(bilateral)

.001 .000 .061 .848 .424 .444

16 16 16 16 16 16 16

Buy button

win rate

Pearson

correlation

.023 - .474 -. 479 1 .285 .702

-. 280

Significance

(bilateral)

.932 .064 .061 .285 .002 .293

16 16 16 16 16 16 16

Table 3: Correlation Analysis.

sales

Click on the

page

Conversion

rate

Page views

Conversion

rate

The buy

button

Win rate

buy

Conversion

rate

Favorable

comment

Conversion

rate

A refund

Conversion

rate

buy

Conversion

rate

Pearson

correlation

.391 -. 057 -. 052 .285 1 .147 -. 053

Significance

(bilateral)

.134 .835 .848 .285 .587 .847

N 16 16 16 16 16 16 16

Favorable

comment

Conversion

rate

Pearson

correlation

.175 - .208 -. 215 .702

.147 1 - .334

Significance

(bilateral)

.518 .440 .424 .002 .587 .207

N 16 16 16 16 16 16 16

A refund

Conversion

rate

Pearson

correlation

278

-. 214 - .206 - .280 - .053 -. 334 1

Significance

(bilateral)

.297 .425 .444 .293 .847 .207

N 16 16 16 16 16 16 16

The Effect of Conversion Rate on Product Sales from the Perspective of AISAS Model: An Empirical Study based on the Data of the

Amazon Merchants

559

As shown in Table 2 and Table 3, five of the

independent variables are positively correlated with

sales volume, which are webpage click conversion

rate, browse conversion rate, purchase conversion

rate, favorable conversion rate and purchase button

winning rate. The increase of the above conversion

rate is conducive to the increase of product sales

volume. However, there is a negative correlation

between the refund conversion rate and the sales

volume, which indicates that the reduction of the

refund conversion rate is helpful to the product sales

volume.

4.2 Regression Analysis

Table 4: Regression Results and Linear Diagnosis of Product Conversion Rate to Sales Volume.

model

Nonstandardized coefficient

The standard

coefficient

t Sig.

Collinearity statistics

Standard error

A trial

version

tolerance VIF

(constant) 341.837 1121.079 - 305. .765

Page click

conversion rate

20623.834 5063.126 .736 4.073 .001 1.000 1.000

(constant) 8071.468 2717.210 -2.970 .011

Page click

conversion rate

21311.505 4038.861 .761 5.277 .000 .997 1.003

Purchase

conversion rate

133199.961 44223.945 .434 3.012 .010 .997 1.003

(constant) 23301.299 6351.225 -3.669 .003

Page click

conversion rate

26037.241 3846.001 .930 6.770 .000 .769 1.301

Purchase

conversion rate

104033.355 38683.153 .339 2.689 .020 .911 1.098

Buy button win

rate

20691.387 8057.829 .367 2.568 .025 .708 1.412

It can be seen from Table 4 that the regression

effect of Model 3 is the best, and the significance

level of t-test is 0.00, 0.020 and 0.025 respectively,

which are all less than 0.05. This indicates that the

relationship between the variable and the dependent

variable - sales volume in Model 3 is significant. The

value of variance inflation coefficient VIF is close to

1, so the multicollinearity in the regression model is

good.

4.3 Results of Hypothesis Test

Analyzing the result of SPSS results, under the

simultaneous action of multiple variables, we can get

the following test conclusions:

i. Hypothesis 1 is valid: the webpage click

conversion rate at the stage of consumers'

interest has a significant impact on product

sales.

ii. Hypothesis 2 is not valid: the browse

conversion rate in the consumer search stage

has no significant effect on the product sales.

iii. Hypothesis 3 is valid: the purchase conversion

rate at the consumer action stage has a

significant impact on product sales.

iv. Hypothesis 4 is not valid: the conversion rate

of favorable comments in consumer sharing

stage has no significant impact on product

sales.

v. Hypothesis 5 is not valid: the refund

conversion rate in the consumer sharing stage

has no significant effect on the product sales

volume.

vi. Hypothesis 6 is valid: the win rate of buy

button has a significant impact on product

sales.

5 MARKETING SUGGESTIONS

5.1 Optimize Keywords and Improve

Product Exposure

The optimization of keywords includes the

optimization of product rankings and advertising (Wu

2018). For the optimization of product ranking,

operators are required to understand the key words

first. Operators need study the search habits of

consumers and potential demand for the products

with the help of some search keywords tools, or filter

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

560

different keywords on foreign shopping websites to

test the importance of products, or select the

appropriate product keywords to improve the product

visibility. The operators can carry out advertising of

the main keywords, optimize the bidding of keywords

according to the situation of advertising, consider

removing the keywords with low transformation, and

bring the maximum quantified potential customers

through keyword optimization.

5.2 Optimize Product Pictures and A+

Pages

In order to improve the shopping experience of

buyers, Amazon has some requirements for product

pictures. The main picture should have a pure white

background, no LOGO, watermark, etc., and the

secondary picture should not contain naked

information or infringement. The standard size of the

picture is 1000*1000 pixels. Under the principle of

the picture system of Amazon, merchants need set the

main picture that meets the requirements and make it

attractive. The auxiliary pictures with multi-function

introduction can be impressed and attracted by

consumers. Beautiful pictures make consumers

interested in the products, then increase the click rate

of webpages, and improve the conversion rate of page

clicks. In addition, the setup of A + page can let

consumers understand product performance, contrast,

size, packaging and application scene. It is easier to

help the consumers understand and use products, to

help consumers make buying decisions and actions in

order to raise the purchase conversion rate of

products.

5.3 Grab the Gold Shopping Cart and

Improve the Conversion Rate of

Commodity Purchase

According to the regulation of the amazon shopping

cart and algorithms, “golden-shopping-cart” is the

unique which is paid bidding for the same product by

many sellers. The sellers who have the listing of new

products, too many negative comments and low

inventory, have no golden-shopping-cart or be

grabbed them by competitors. The sellers who have

no the golden-shopping-cart cannot sell the products

to consumers or create the orders. At the same time,

golden-shopping-cart is also a kind of affirmation to

high-quality sellers. Consumers will give priority to

the sellers with golden-shopping-cart. Therefore,

continuously grabbing golden-shopping-cart and

obtaining a competitive turnover rate can improve the

conversion rate of purchase and increase the sales

volume.

5.4 Pay Attention to Consumers'

Sharing and Improve Consumer

Experience

Amazon pays close attention to buyers' comments on

products. Positive comments can help other

consumers make decisions in online purchasing

behavior, and also improve product ranking, purchase

conversion rate and sales volume. Although the

refund conversion rate cannot be recognized by

consumers, it is an important factor to measure the

negative experience of customers. For foreign

consumers, they pay attention to the experience. The

sharing about the quality of products and the

experience of using the product is the key to lead

consumers to make decisions. When consumers have

a comfortable experience of the product, consumers'

sharing can influence other consumers to make

purchase decisions, and also help CBEC enterprises

to establish a product promotion system (Feng 2016).

6 CONCLUSIONS

This paper studies the influence of the conversion rate

of goods at different stages on the sales volume of

Amazon online stores, carries out an empirical test to

understand the relationship between variables, and

draws the final conclusion. However, there are still

some deficiencies in this paper that need to be

improved and perfected. Firstly, due to the limited

product samples, the research on commodity

conversion rate has certain limitations. More products

can be studied in the future. Secondly, the sample

data selected for the study is the data information of

the company's products within two months, which

can be studied and verified for a longer period of

product data in the future.

ACKNOWLEDGMENTS

This paper is supported by the project of the Ministry

of education of P. R. China under grant No.

201901226017 and No. 201801193068, the project of

Guangdong Province Education Department under

grant No. 2019J054 and No. 2017WQNCX187.

The Effect of Conversion Rate on Product Sales from the Perspective of AISAS Model: An Empirical Study based on the Data of the

Amazon Merchants

561

REFERENCES

Bao-Wu Bian, Wen-Hua Dong (2009). Research on

Commodity Attraction Factors Affecting the

Conversion Rate of Enterprise E-Commerce Website

[J]. E-commerce,2009 (07) :74-77.

Guang-Ting Zhu (2015). Research on Consumer Behavior

in C2C Mode under Big Data Environment [D].

Shanghai University of Engineering Science.

Jiang Jing, Zhi-Yong Yang (2013). An Empirical Study on

the Sales Conversion Rate of Product Attention and

Purchase Intention Based on Purchase Behavior Pattern

[J]. Shanghai Management Science,2013(08):25-30.

Kuester Sabine (2012). Strategic Marketing & Marketing in

Specific Industry Contexts, University of

Mannheim:110.

Na Zhou, Nian-Nian Liu, Shan Li (2017). Research on

Influencing Factors of Online Consumers' Buying

Decision Based on R-Type System Clustering Analysis

[J]. China Market,2017(14):40-41+53.

Tan Kai Yee (2018). Research on Commodity Conversion

Rate and Its Impact on Cross-border E-commerce

Third-Party Trading Platform [D]. Zhejiang University.

Xi-Ling Wu, Chi-Ping Yuan (2018). Cross-border e-

commerce practices [M]. Tsinghua University Press.

Ying-Jian Feng (2016). Network marketing foundation and

practice[M]. Tsinghua University Press.

Zan Mo, Yan-Fei Li, Peng Fan (2015). Online Reviews of

Consumer Purchase Behaviors [J]. Journal of Service

Science and Management,8(3):419-424.

Zhi-Cheng Li, Mei-Lian Liu (2002). Research on consumer

behavior under e-commerce environment [J]. China

Management Science,2002(06):89-92.

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

562