The Impact of Instagram “Call-to-Action” Buttons on

Customers’ Impulse Buying

Regina C. Handayani Betty Purwandari
Faculty of Computer Science Faculty of Computer Science
Universitas Indonesia Universitas Indonesia
Depok, Indonesia Depok, Indonesia
regina.carla@ui.ac.id bettyp@cs.ui.ac.id

Iis Solichah Pudy Prima

Faculty of Computer Science Faculty of Computer Science
Universitas Indonesia Universitas Indonesia
Depok, Indonesia Depok, Indonesia
iis@cs.ui.ac.id pudy.prima61@ui.ac.id

ABSTRACT platform for e-commerce [22]. Moreover, the world’s Internet

High penetration of social media in Indonesia gives more penetration reached 50% of the total population in January 2017
opportunities for sellers to promote their products, as well as for [6]. Most users accessed the Internet on mobile devices. There
consumers to buy. Instagram is a popular social media for e- were 40% penetration from total population and 106 million
commerce in Indonesia. In 2016 Instagram introduced the Call-to- active social media users in Indonesia [6].
Action (CTA) buttons to trigger more e-commerce transactions on In early 2017, Instagram engagement rate reached 70% higher
Instagram. On the other hand, impulse buying of social commerce than Facebook [10]. Since the inclusion of e-commerce features
consumers can be explained by the Stimulus-Organism-Response into Instagram, many industries use it as a platform to advertise
(S-O-R) model. Therefore, the objective of this study is to their products or services in increasing sales and brand awareness
investigate the impact of the CTA buttons on consumers’ impulse [12] [26]. In early 2018, Indonesia was the third countries with the
buying using the S-O-R model. A questionnaire was administered largest active Instagram user [24]. It especially applied to teens
to collect data from 163 consumers. The data were analyzed and and young people, who intensively interacted with social media [2]
tested using Partial Least Squares (PLS). The results were [5]. Thus, Instagram is a popular social media among e-commerce
validated by interviews with 3 sellers and 3 consumers. It shows in Indonesia [11].
that the CTA buttons have a positive impact on consumers’
impulsive purchase, but less effective to improve consumers’ Showing photos or videos of products on sellers’ Instagram
perceived enjoyment. The main contribution of this study is the accounts become one of the most important factors for
understanding of consumers' impulse buying response on encouraging impulse buying [15]. In 2016 Instagram released its
promotional ads feature, which is related to the CTA buttons. CTA buttons in their promotional ads feature [9]. This feature is
set up to help online stores to brand and share their products or
CCS Concepts services on Instagram. In return, Instagram earns revenue from
• Information systems ➝ World Wide Web ➝ Web online shops using promotional advertising features. As a result,
applications ➝ Electronic commerce ➝Online shopping using CTA buttons makes advertisement product in Instagram
more attractive and trigger potential customers to be more
Keywords impulsive in buying.
E-commerce; Social Commerce; Impulse Buying; Instagram for On the other hand, there are previous studies on how the S-O-R
Business; Partial Least Squares; Call-to-Action model explained impulse buying on Facebook [13]. It leads to this
study on the use of the model to understand impulse buying on
1. INTRODUCTION Instagram. Therefore, the aim of this research is to investigate the
Mobile commerce combined with social commerce is a powerful impact of Instagram for business promotional advertising features
Permission to make digital or hard copies of all or part of this work for and CTA buttons on consumers’ impulse purchase using S-O-R
personal or classroom use is granted without fee provided that copies are model.
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. Copyrights for 2. LITERATURE REVIEW
components of this work owned by others than ACM must be honored. Some research has conducted literature studies on social media
Abstracting with credit is permitted. To copy otherwise, or republish, to from various perspectives [3] [21] [20]. They found that social
post on servers or to redistribute to lists, requires prior specific permission commerce is a cross-breeding of the social media with the
and/or a fee. Request permissions from Permissions@acm.org.
electronic commerce involving users as sellers or buyers and
ICBIM’18, September 20–22, 2018, Barcelona, Spain.
© 2018 Association for Computing Machinery marketing it on the Web 2.0 [16] [12].
ACM ISBN 978-1-4503-6545-1/18/09 …$15.00.
DOI: http://doi.org/10.1145/3278252.3278276
50
2.1 The Emergence of Instagram Commerce constructs. PLS-SEM can be used to test hypotheses using
First appeared as a social media for sharing photos, now statistical models developed from relevant theories [18] [25].
Instagram offers a business tool to visualize advertisements and
market products more easily. Instagram business profile contains
3.1 Hypotheses Development
promotional ads buttons and analytics features that can be utilized Based on previous discussions about S-O-R model can explained
by online stores to market their products to potential customers. impulse buying on Facebook, we start to explore impulse buying
on Instagram for business with CTA buttons using S-O-R model.
There are several features added by Instagram to differentiate a
personal profile from a business profile [17]. It provides detailed Stimulus
information on the followers’. The information is presented in the Product
Response
Availability
form of demographic of followers, profile impressions, profile Organism
access time, and profile views. This is the same strategy as H1a
H5a
Impulsiveness

Facebook, which enables users to organize, run, and track their H2a
Perceived
Usefulness
advertising campaigns [12]. Visual Appeal
H4 H5b H7
H2b
2.2 Impulse Buying H1b
H6a

Literature review on impulse buying reveals that the antecedents Perceived

H3a Enjoyment
of impulse buying consists of three main factors: (1) Dispositional Ease of Use H6b
Urge to Buy
factors, which includes Impulse Buying Trait (IBT), H3b
Impulsively

psychographics, and dispositional motivation; (2) Situational Advertisement

factors derived from social influence, affect, constraints such as Content

time or available money, external cues, retail environments,

shopping behavior like browsing, situational motivation, or Figure 1. Research odel
product characteristics; (3) Sociodemographic factors that include
gender, age, ethnicity, and income [1]. Therefore, the research model and the following hypotheses are
proposed:
There are four types of online impulse purchase in the literature.
First, when consumers buy products at certain conditions, but they Instagram provides visual sharing on photos and videos with
have planned for it beforehand. Second, pure impulse purchase, various content which supports multiple photo in one-time post.
when consumers buy products or services unplanned. Third, Consumers are more interested in online stores which display a
impulsive purchase, because they are reminded of those products variety of products visually in their catalog [14]. Based on the
displayed in stores or advertisements and having the urge to buy above explanation, we compose a hypothesis:
those products in the past. Last, impulse purchase when H1a Perceived product availability of promotional features
consumers buy products because of some suggestions from positively affects perceived visual appeal.
friends, family or closest people [15].
Previous research showed the relation between perceived ease of
2.3 Stimulus-Organism-Response (S-O-R) use Website [23] and visual appeal [14]. It was proven that an
Model easy-to-use Website increased impulsive purchase from the
The classic S-R model was first introduced in 1929 by consumers. We adopted this into our hypothesis. We take the
Woodworth, but as the most appropriate theoretical background CTA buttons on promotional ad buttons developed by Instagram
for the concept in the field of psychology studies is the extension appear on the potential consumer timeline. Then, there are options
of the classical S-O-R model [4]. of “Shop Now” or “Visit Profile” from an online store to make it
easy. Therefore, another hypothesis is composed:
This research adopts the S-O-R model as a corridor in the
hypothetical planning. The fundamental of human behavior in the H1b Perceived ease of use of promotional features positively
stimulus is attitude, motivation, thought, or emotion, which is affects perceived visual appeal.
used by the organism as a tool to estimate the response to be The first impression of an online store is very influential on user
generated [19]. activity in the future. The perception that the visual elements of
The S-O-R model proven can explained consumer impulse buying the promotional ads feature can elicit usefulness and enjoyment
because of this model is quite effective in influencing persuasion for Instagram users [28]. Therefore, we compose two other
due to the intensity of communication between individuals or hypotheses:
certain groups on Facebook [13]. H2a Perceived visual appeal of promotional features positively
affects perceived usefulness.
3. RESEARCH METHODOLOGY
We define 1) Stimulus as the Instagram features, where H2b Perceived visual appeal of promotional features positively
advertisements can be inserted for business purposes; 2) Organism affects perceived enjoyment.
as the Instagram users, who have experienced at least one Some recent research in the literature discussed that information
purchasing transaction on Instagram; and 3) Response as the on online stores should be communicated to the consumers to
reaction of the Organism given the Stimulus. The research model
increase customer returns to the shopping site. Improving the
is shown in Figure 1. quality of information from an advertising content is predicted to
In this study, data was statistically analyzed using Partial Least help the consumers to receive perceived usefulness and enjoyment
Square (PLS)-Structural equations modeling (SEM), which aims [7]. Thus, the following hypotheses are inferred:
to investigate the relationship between endogenous and exogenous

51
H3a Perceived advertisement content on promotional features and then excluded. Therefore, the sample size used for this study
positively affects perceived usefulness. was 163. All of them were Instagram consumers.
H3b Perceived advertisement content of promotional features The demographic data of all respondents are listed in Table 1.
positively affects perceived enjoyment. There were more female respondents (68.1%) than men (31.9%).
Most respondents lived in DKI Jakarta (68.1%). Teenagers and
This study would like to investigate whether the features on young adults between 17-25 years old (77.3%) were majority of
Instagram provide cognitive reactions to individuals. If the participants. There were 42.3% of respondents earning 1-5
individuals feel something useful from their shopping activities on million Indonesian Rupiah (IDR) or 70-350 USD per month. Most
Instagram, it means that they get the convenience of shopping participants accessed Instagram several times a day (84.7%), and
experience [28]. Another hypothesis is proposed: experienced Instagram shopping less than 5 times per month
H4 Perceived usefulness of promotional features positively affects (65%).
perceived enjoyment.
4.2 Measurement Evaluation
The emotional response of the individual becomes one of the In testing the model, structural equation modeling was employed
benchmarks for the next action of the users [7]. As the case of to investigate the relationship between endogenous and exogenous
online shopping behavior [27], Instagram provides the large constructs. The first analysis was established by measuring the
selection of stores, variety of products, as well as reviews from validity and reliability of each latent variable by using SmartPLS
previous customers, which can affect the mood of consumers. software. Reflective measurement of an indicator will be valid if it
Their good mood with the site can trigger perceived usefulness has a loading value (λ) above 0.5, if one indicator has a value
and impulse buying. Two other hypotheses are proposed: below 0.5 then the indicator must be discarded because it
H5a User perceived usefulness of promotional features positively indicates that the indicator is not good enough to measure the
affects user impulsiveness. latent variable appropriately. In this study all reflective indicators
have a loading.
H5b User perceived usefulness of promotional features positively
affects user urge to buy impulsively. Table 1. Respondents Demographic Data

Every individual would expect a satisfaction from a purchase they Percentage

Characteristics Frequency
have made, their positive mood while searching for a product or (100%)
service online is very influential [23] [8]. In this study, we predict Female 68.1 111
that promotional features on Instagram can increase impulsiveness Gender
Male 31.9 52
of the consumers as well as their urge to make a purchase [28].
Therefore, we propose the following hypotheses: DKI Jakarta 68.1 111
Java 24.5 40
H6a Perceived enjoyment of promotional features positively
affects user impulsiveness. Location Sumatera 4.9 8
Borneo 1.8 3
H6b Perceived enjoyment of promotional features positively
affects user urge to buy impulsively. East Nusa Tenggara 0.6 1
< 17 3.7 6
Previous research states when individuals have given a positive
response [14] then they tend to buy impulsively, and most likely 17 – 25 77.3 126
Age (Years)
have the impulse to make urgent purchase. On Instagram, all 26 – 45 17.8 29
convenience offered by the CTA buttons might trigger the users’ > 45 1.2 2
urge to buy impulsively. Thus, the following hypothesis is
proposed: < 1 million 34.4 56
Income 1 – 5 million 42.3 69
H7 Perceived impulsiveness of promotional features positively (IDR/
affects users’ urge to buy impulsively. Month) > 5 – 10 million 20.2 33
> 10 million 3.1 5
3.2 Questionnaire design and measurement Several times a day 84.7 138
Scales Once a day 8.0 13
This research performs quantitative methods to test above
The Several times a
hypotheses. The data were collected from using questionnaires. 4.3 7
frequency of week
The constructs, indicators, and questions for hypothesis testing
Instagram Once a week 1.2 2
were developed from a comprehensive literature study [14] [28]
Access Several times a
[7]. It was then discussed by e-commerce experts for validation. 0.6 1
This study used Likert scale measurement with a five-point scale month
from 1 (strongly disagree) to 5 (strongly agree). Several times a year 1.2 2
Never use before 23.3 38
4. RESULT AND ANALYSIS Instagram
Shopping < 5 times 65.0 106
4.1 Demographic Data
In this study, the questionnaire was distributed online in Experience 5 – 8 times 5.5 9
(per Month) > 8 times
December 2017 to Instagram users, who have used this platform 6.1 10
to search for a product or service. There were 165 responses n=163
collected within ten days. However, 2 of them were incomplete
52
 who previously had no intention to buy become interested and feel
the desire to buy the product
value (λ) above 0.5, so all indicators are valid and none is
removed. All the constructs have met the requirements that the path
coefficient value higher than 1.97. This suggests that all the
The outer model analysis is to use unidimensionality to measure remaining hypotheses are accepted with positive direction. The
the reliability of the model. The measure is conducted to first rejected hypothesis on the first analysis attempt (H3b), shows
determine the composite reliability and Cronbach’s alpha value. that Advertisement Content did not directly affect perceived
Table 2. Reliability and Validity enjoyment.

Composite Table 3. Path Coefficient Result

Cronbach’s
Variables Reliability AVE R2 Hypo-
Path
T Statistics Status
Alpha thesis (|O/STDEV|)
(PLS)
H1a PA → VA 0.477 Accepted
PA 0.875 0.923 0.801 N/A
H1b EOU → VA 0.187 Accepted
VA 0.813 0.813 0.727 0.362 H2a VA → PU 0.395 Accepted
EOU 0.848 0.908 0.766 N/A H2b VA → PE 0.429 Accepted
AC 0.749 0.888 0.799 N/A H3a AC → PU 0.281 Accepted
PU 0.939 0.961 0.891 0.368 H3b AC → PE 0.109 Rejected
H4 PU → PE 0.352 Accepted
PE 0.959 0.973 0.924 0.478
H5a PU → IMP 0.416 Accepted
IMP 0.845 0.896 0.684 0.385 H5b PU → UBI 0.215 Accepted
UBI 0.940 0.962 0.894 0.433 H6a PE → IMP 0.276 Accepted
The cut-off value of the composite reliability is 0.7 to indicate an H6b PE → UBI 0.006 Rejected
appropriate internal stability for the model and for Cronbach’s H7 IMP → UBI 0.510 Accepted
alpha is 0.6 [8].
An explanation for this would be: The more advertisements
As shown in Table 2, the composite reliability and Cronbach’s appear on the sidelines of Instagram pages the more
alpha values are above the cut-off value. These results can inconvenience experienced by the users. The second rejected
confirm the model reliability. Furthermore, the next step in hypothesis (H6b), shows that it is not always the case that if the
examining the validity model is by using Average Variance customer is comfortable then they are encouraged to make urgent
Extended (AVE), which can represent the correlation between buying.
variables and index. The higher correlation value will have a good
effect. The critical value of AVE is 0.4 if greater then it will be 5. FURTHER INTERVIEW
accepted [8]. A follow up interview was conducted after PLS analysis to
explore explanations and implications of the survey results.
4.3 Hypotheses Testing Interviews were conducted with 6 Instagram users by phones or
For the testing phase and confirmation of hypothesis in this study, face-to-face meetings. Half of the users were seller from online
the existing data will be tested based on two criteria that are R 2 shop, and the other half were consumers. The interviewees’
and significance value using t-test. The use of R2 criterion aims to profiles are summarized in Table 4 and 5 represent consumer and
connect the structural part of modeling equations and online seller.
measurements, to reveal the effect of independent variables on the
From the perspective of consumers who have done online
dependent variable [8].
shopping on Instagram, Person 1 and 2 stated that promotional ads
Then, the result shows that the PE construct gives the highest R 2 on the feeds page did not have significant effect on their impulse
value of 0,478, and second highest is the UBI with the value of buying. Meanwhile, Person 3, who is younger, said that
0,433. Afterward, we conducted bootstrapping to observe the t- promotional ads on Instagram have increased her frequency of
statistic values in path coefficient, which must be higher than 1.97 impulse purchase.
and reside at the 0.025 (2.5%) significance level. Hypotheses that
Although ads features on Instagram gave less enjoyment and
cannot meet the path coefficient value will be eliminated.
pleasure, the consumers admitted promotional ads could add their
The t-test result revealed almost all hypotheses acceptable (H1a, insight to a new product or service. They also agreed that the first
H1b, H2a, H2b, H3a, H4, H5a, H5b, H6a, H7) but there are two aspect of their concern was the visual appeal of an online store.
hypotheses that are less than 1.97, i.e. H3b (AC → PE) and H6b
Furthermore, an online seller who has switched to Instagram for
(PE → UBI). It means that those hypotheses are rejected and business, found out more about their potential customers using
should be removed. Correlation coefficient values for each of the existing analytics on insight feature. Promote ads helped her to
two dimensions can be seen in the Table 3. broadcast products or services on Instagram. Consequently, it
The findings of this study indicate that both perceived usefulness increased the number of visits to the online stores.
and perceived enjoyment. Promotional ads content on Instagram The main constraint is Instagram requirement for payment of its
can affect the impulsiveness of the users and that may trigger promotional ads features only by credit card only. The outcomes
impulse buying. The attractive advertisement content on the of follow-up interviews echoed similarities with survey results. It
timeline on Instagram in terms of visualization can make users revealed that only a small proportion of consumers were happy

53
with the presence of promotional ads on their Instagram feeds, 7. CONCLUSION
because most of them are indifferent to this promotion. Consumer acceptance of advertising campaigns using Call-to-
Table 4. Consumer Profiles Action (CTA) buttons in certain age groups only affects their
purchase intentions and has not yet reached impulse buying from
Characteristics Person 1 Person 2 Person 3 consumers. However, in the younger age group of Instagram
Gender Male Female Female promotional ads resulted in their impulse buying trends in online
Age 24 23 19 stores. The results show that the responses estimated with the
Occupation Banking Taxing Student stimulus of visual attractiveness and advertising content based on
Domicile Jakarta Jakarta Batam the S-O-R model should be given special attention.
Instagram shopping 3 years 4 years 3 years Hence, the experience of shopping at an attractive online store
experience visually can improve a positive mood that will result in a tendency
Marketplace 2 years 1 years 1 years to buy products impulsively. The advertising content displayed by
shopping experience Instagram to its users in the timeline can make it easier for them
to get information about the appropriate product.

Table 5. Online Seller Profiles This research can provide new insights on understanding the
reasons for consumer decision making before making a purchase.
Characteristics Seller 1 Seller 2 Seller 3 In addition, promotional advertising features on Instagram can be
Role Owner Owner Owner used as a competitive strategy by companies or industries with
Gender Female Female Male limited initial capital. Promotional ads on Instagram offer
Age 18 18 27 relatively cheaper cost, can increasing customer awareness and
Degree Under- Under- Post- impulse buy.
graduate graduate graduate
Business type Retail Retail Retail 8. LIMITATION AND FUTURE
Business 1 – 5 years 1 – 5 1 – 5 RESEARCH
running years years The trust issue on social commerce has not been discussed in this
Employee < 10 people < 10 < 10 study. In addition, most of survey respondents and interviewees
people people represent the younger generation so cannot be generalized. This
Instagram Yes Yes Yes study investigated only promotional advertisements on Instagram
business timelines containing Call-to-Action (CTA) buttons feature
Offline store No No No excluding Instagram stories with a swipe up feature that may also
Other None None Yes affect consumers’ impulse buying. Future research is expected to
marketplace analyze more about the factors that can increase the trust of
consumers on Instagram of online stores that have switched using
6. DISCUSSION Instagram for business.
The results of the confirmed hypothesis show that almost all the
stimuli given an affect the organism in generating an impulse 9. ACKNOWLEDGMENTS
buying response. Results relating to the previous t-test revealed This study is funded by a PITTA Grant No.
that the stimuli provided by visual appeal can lead to perceived 1869/UN2.R3.1/HKP.05.00/2018 from the Universitas Indonesia.
usefulness and enjoyment. Whereas, advertisement content only
affects perceived usefulness. These two perceptions can produce APPENDIX A: Measurement Items
an impulsiveness response that leads to impulse buying. Product Availability [14]
There are two implications in this research for online stores on 1. Promoted ads on Instagram shows the variety of products
Instagram business profiles. Firstly, now Instagram begin to available to me
transform into a social commerce platform and Call-to-Action 2. Promoted ads on Instagram can easily find what I need
(CTA) buttons feature on promotional ads can help online stores 3. Promoted ads on Instagram can easily find what I am
increasing their sales by broadcasting an advertisement to an interested in
audience that does not even follow their online store account.
Visual Appeal [14]
Although most of the users' response is indifferent to promotional 1. Promoted ads on Instagram are visually pleasing
advertising features on Instagram, there is still an opportunity for 2. Promoted ads on Instagram display visually pleasing
online shop owners to get the attention of potential customers. designs
They can advertise their product by uploading photos or videos 3. The placement of promoted ads on Instagram are attractive
that are visually appealing, paying attention to the time when
uploading them. Ease of Use [14]
1. The information promoted advertising on Instagram can be
Promotional ads on Instagram still not be able to give users'
obtained in a simple way
purchase satisfaction, it’s because after the user click the CTA
2. The advertising information promoted on Instagram can be
buttons, there are some other processes to go through before they
obtained easily
successfully buy the product they want. Hence, perceived
3. Ads promoted on Instagram display relevant information
enjoyment cannot directly affect consumers’ urge to buy
obvious and easy to find
impulsively.
Advertisement Content [7]
54
1. Promoted ads on Instagram provide detailed information [8] Hair, J.F. et al. 2016. A Primer on Partial Least Squares
that is relevant to consumer needs Structural Equation Modeling (PLS-SEM). Sage Publishing.
2. Promoted ads on Instagram shows visual product [9] Instagram Ads Call to Action: 2016.
information of good quality https://business.instagram.com/advertising. Accessed: 2017-
Perceived Usefulness [28] 11-20.
1. Promoted ads on Instagram can improve my shopping [10] Instagram Engagement Rate Is 70% Higher Than Facebook’s:
performance 2017. https://locowise.com/blog/instagram-engagement-rate-
2. Promoted ads on Instagram can improve my shopping is-higher-than-facebook. Accessed: 2017-11-27.
productivity [11] Jakarta Post 2017. Indonesia Instagram’s Biggest Market in
3. Promoted ads on Instagram can increase my shopping Asia Pasific. Science and Tech.
effectiveness
[12] Laudon, K.C. and Traver, C.G. 2017. E-commerce 2017
Perceived Enjoyment [28] Business, Technology, Society. Pearson.
1. The emergence of promoted ads on Instagram was [13] Leong, L.Y. et al. 2018. The effects of Facebook browsing
enjoyable and usage intensity on impulse purchase in f-commerce.
2. The emergence of promoted ads on Instagram was exciting Computers in Human Behavior. 78, (2018), 160–173.
3. The emergence of promoted ads on Instagram was pleasant DOI:https://doi.org/10.1016/j.chb.2017.09.033.
Impulsiveness [7] [14] Liu, Y. et al. 2013. Website attributes in urging online
1. When I shop online on Instagram, I buy things that I had not impulse purchase: An empirical investigation on consumer
intended to purchase perceptions. Decision Support Systems. 55, 3 (2013), 829–
2. I often buy things on Instagram without considering the 837. DOI:https://doi.org/10.1016/j.dss.2013.04.001.
consequences [15] Lo, L.Y.S. et al. 2016. Motivation for online impulse buying:
3. When I see something that really interests me on Instagram, I A two-factor theory perspective. International Journal of
simply buy it Information Management. 36, 5 (2016), 759–772.
4. Buying online on Instagram spontaneously is fun DOI:https://doi.org/10.1016/j.ijinfomgt.2016.04.012.
Urge to Buy Impulsively [28] [16] Mata, F.J. and Quesada, A. 2014. Web 2.0, social networks
1. When I saw promoted ads, I had the urge to purchase items and E-commerce as marketing tools. Journal of Theoretical
that are not my shopping goal and Applied Electronic Commerce Research. 9, 1 (2014).
2. When I saw promoted ads, I had desire to buy items that did DOI:https://doi.org/10.4067/S0718-18762014000100006.
not pertain to my shopping goal [17] Personal Instagram vs. Instagram Business Profile: 2017.
3. When looking at promoted ads, I had the inclination to https://later.com/blog/instagram-business-profile/. Accessed:
purchase items outside my shopping goal 2017-11-21.
[18] Purwantoro, F. et al. 2016. E-audit system acceptance in the
10. REFERENCES public sector: An Indonesian perspective. ICACSIS 2015 -
[1] Amos, C. et al. 2014. A meta-analysis of consumer impulse
2015 International Conference on Advanced Computer
buying. Journal of Retailing and Consumer Services. 21, 2
Science and Information Systems, Proceedings (2016).
(2014), 86–97.
DOI:https://doi.org/10.1016/j.jretconser.2013.11.004. [19] Rajaguru, R. 2014. Motion Picture-Induced Visual, Vocal
and Celebrity Effects on Tourism Motivation: Stimulus
[2] APJII 2016. Penetrasi Survei Internet & Perilaku Pengguna.
Organism Response Model. Asia Pacific Journal of Tourism
(2016).
Research. 19, 4 (2014), 375–388.
[3] Busalim, A.H. and Hussin, A.R.C. 2016. Understanding DOI:https://doi.org/10.1080/10941665.2013.764337.
social commerce: A systematic literature review and
[20] Salvatori, L. and Marcantoni, F. 2015. Social Commerce : A
directions for further research. International Journal of
Literature Review. Science and Information Conference
Information Management. 36, 6 (2016), 1075–1088.
2015. (2015), 1–6.
DOI:https://doi.org/10.1016/j.ijinfomgt.2016.06.005.
DOI:https://doi.org/10.1109/SAI.2015.7237152.
[4] Buxbaum, O. 2016. Key Insights into Basic Mechanisms of
[21] Shanmugam, M. and Jusoh, Y.Y. 2014. Social commerce
Mental Activity.
from the Information Systems perspective: A systematic
[5] Dey, D.K. and Srivastava, A. 2017. Impulse buying literature review. International Conference on Computer and
intentions of young consumers from a hedonic shopping Information Sciences (ICCOINS). (2014), 1–6.
perspective. Journal of Indian Business Research. (2017), DOI:https://doi.org/10.1109/ICCOINS.2014.6868435.
JIBR-02-2017-0018. DOI:https://doi.org/10.1108/JIBR-02-
[22] Turban, E. et al. 2018. Electronic Commerce 2018 A
2017-0018.
Managerial and Social Networks Perspective. Springer
[6] Digtal in 2017 Global Review: 2017. International Publishing AG.
https://wearesocial.com/sg/blog/2017/01/digital-in-2017-
[23] Verhagen, T. and Van Dolen, W. 2011. The influence of
global-overview. Accessed: 2017-09-29.
online store beliefs on consumer online impulse buying: A
[7] Floh, A. and Madlberger, M. 2013. The role of atmospheric model and empirical application. Information and
cues in online impulse-buying behavior. Electronic Management. 48, 8 (2011), 320–327.
Commerce Research and Applications. 12, 6 (2013), 425– DOI:https://doi.org/10.1016/j.im.2011.08.001.
439. DOI:https://doi.org/10.1016/j.elerap.2013.06.001.
55
[24] We Are Social and Hootsuite 2018. Digital in 2018 Report. Electrical Engineering, ICITACEE 2016. 1975 (2017), 1–6.
30/01/2018. (2018). DOI:https://doi.org/10.1109/ICITACEE.2016.7892472.
[25] Wijaya, K. et al. 2016. Personal traits as antecedents towards [27] Wu, I.L. et al. 2016. Defining key drivers of online impulse
intention to use: A perspective of a government EDMS purchasing: A perspective of both impulse shoppers and
adoption in Indonesia. ICACSIS 2015 - 2015 International system users. International Journal of Information
Conference on Advanced Computer Science and Information Management. 36, 3 (2016), 284–296.
Systems, Proceedings (2016). DOI:https://doi.org/10.1016/j.ijinfomgt.2015.11.015.
[26] Wiradinata, T. and Iswandi, B. 2017. The analysis of [28] Xiang, L. et al. 2016. Exploring consumers’ impulse buying
Instagram Technology adoption as marketing tools by small behavior on social commerce platform: The role of
medium enterprise. Proceedings - 2016 3rd International parasocial interaction. International Journal of Information
Conference on Information Technology, Computer, and Management. 36, 3 (2016), 333–347.
DOI:https://doi.org/10.1016/j.ijinfomgt.2015.11.002.

56