Omnichannel Shopping Intention in Indonesian Online-to-Offline Grocery Retailers

. With the change in shopping trends, grocery retailers, which were previously present in offline channels, began to expand by also presenting online channels for their customers, resulting in the emergence of the online-to-offline (O2O) phenomenon for groceries retail that applies the omnichannel concept in order to create an omnichannel customer experience and resulting in omnichannel shopping intention through perceived compatibility and perceived risk. This research was conducted on 251 respondents through an online questionnaire using purposive sampling, and the research data was analysed using the SEM-PLS method. This study concludes that Connectivity, Integration, and Consistency as elements of omnichannel customer experience influence Perceived Compatibility; Personalization also influences Perceived Risk; then Perceived Compatibility influences Omnichannel Shopping Intention, while Perceived Risk has no significant effect on Omnichannel Shopping Intention. The results of this study are expected to provide insight for the O2O grocery retailers in Indonesia regarding customer experience to increase their shopping intention in an omnichannel context.


Introduction
The growth of online grocery in Asia from 2019 to 2023 is expected to reach 198% [1].This is also the case in Indonesia, where 1.2% of customers have shopped for food online in 2016 and this figure is expected to continue to grow to 5.4% by 2030 [2].Industry players also agree with this statement and assume that the shopping trend of Indonesian customers has shifted to online grocery, where customer demand for an easy and fast shopping experience is increasing [3].Online grocery shopping allows retailers to be able to connect with potential customers 24x7 and create a dynamic virtual store to facilitate product purchases, customer service, and get product information anytime [4].
Seeing this change in shopping trends, grocery retailers which were previously available in offline channels, began to expand by also presenting online channels for their customers, resulting in the phenomenon of online-to-offline (O2O) groceries retail that applies the omnichannel principle.Omnichannel is a unified approach that manages channels as intermingled touch points to allow consumers to have a seamless experience within an ecosystem [5].Of course, the application of the omnichannel principle is a challenge for industry players because it does not only offer more than one sales channel to customers, but also integrates the various sales channels used.As stated by Hoogveld et al. [6] that the evolution to omnichannel retailing represents a big challenge for retailers who must modify their business models to integrate their * Corresponding author: brian_muchardie@binus.ac.id core-sources by analysing how customers use digital and physical touchpoints simultaneously to make purchase decisions.
For grocery retailers in Indonesia who are in a trial period in implementing omnichannel in their business, it is important to know how consumers' omnichannel shopping intentions are generated.Talking about omnichannel shopping intention, Nielsen stated that from 2016 to 2018 there was an increase in the percentage of customers who like shopping omnichannel from 55% to 66% [7].Moreover, the results of research conducted by Setiawan et al. [8] stated that nearly 40% of customers in Indonesia claimed to be implementing a pattern of shopping behaviour to become more omniscient during the post-Covid-19 pandemic, where they searched for products online but made transactions offline and otherwise.This shows that there has been omnichannel shopping intention among customers in Indonesia.To confirm this fact, preliminary research was conducted on 70 respondents, and it was found that 72.6% of respondents chose the integration of online and offline channels as when shopping for groceries, which refers to the omnichannel concept.
In understanding how to make customers have omnichannel shopping intention, it is necessary to examine the influence of the main factors that drive it.Agag et al. [9] used innovation diffusion theory as a theoretical framework in examining the factors that influence customer intentions to adopt online travel purchasing methods.Following this framework, Shi et al. [10] conducted a qualitative study to determine the factors that influence omnichannel shopping intention.The result is, omnichannel customer experience which consists of connectivity, integration, consistency, flexibility, and personalization is known to be an antecedent of omnichannel shopping intention.Customer experience is considered as an interactive holistic process, facilitated through cognitive and emotional cues, moderated by the customer and contextual characteristics so as to produce a unique and pleasant or unpleasant experience for the customer [11].In the omnichannel context, Neslin et al. [12] and Verhoef et al. [13] add that although the results of research on information and communication technology in the omnichannel context are increasing, it is important to continue to apply investigations to the customer behaviour area in this area in an effort to find various insights on how to create seamless customer experience through omnichannel.
Omnichannel customer experience as an antecedent of omnichannel shopping intention goes through a process before finally being able to influence omnichannel shopping intention.Shi et al. [10] proposes perceived compatibility and perceived risk which are characteristics of innovation as the basic mechanism that must be passed by omnichannel customer experience in the process of generating omnichannel shopping intention.This is in line with Dwivedi et al. [14] which states that individuals make decisions to adopt or reject an innovation based on their perceptions and beliefs about the innovation.Ozturk et al. [15] added that perceived compatibility is the degree or value in which an innovation is considered consistent with the values, previous experiences, and needs of potential consumers.When customers switch channels from offline to online and vice versa, the perceived compatibility that is felt in both channels will determine their intention to buy [16].Meanwhile, perceived risk is a risk perceived by customers related to their concerns that the channels used are not in line with their expectations [17].Therefore, industry players need to convince customers through the given omnichannel customer experience that the innovation in the form of the omnichannel concept offered is able to provide compatibility and reduce risk for customers.
H1: Perceived Compatibility significantly affects Omnichannel Shopping Intention in O2O grocery retailers.
H2: Perceived Risk significantly affects Omnichannel Shopping Intention in O2O grocery retailers.
The concept of connectivity in the omnichannel context is the extent to which customers frequently switch from one channel to another, expecting the omnichannel retailer to guide directions and facilitate smooth transitions across all channels [10].Dwivendi et al. [14] gave an example where a mobile application can provide directions and marketing information from the nearest physical store for customers to try products, which in turn result in perceived compatibility obtained by customers.H3: Connectivity significantly affects Perceived Compatibility in O2O grocery retailers.
Verhoef et al. [13] mentioned that integration must be considered as an important factor in the context of omnichannel shopping.Shi et al. [10] justified that statement by arguing that integration is the extent to which the customer perceives all information systems and service content are unified and integrated well across channels.
H4: Integration significantly affects Perceived Compatibility in O2O grocery retailers.Consistency is the extent to which customers experience both content and process consistency across channels [10].The consistency of cross-channel information at the same time will have a synergy effect and will provide continuity of service when consumers switch to other channels, resulting in minimal performance risk as long as consumers shop on omnichannel [18].
H5: Consistency significantly affects Perceived Compatibility in O2O grocery retailers.
H6: Consistency significantly affects Perceived Risk in O2O grocery retailers.
Flexibility is an effort done by retailers to provide flexible options and experience to consumers to be able to switch to other channels, where services are readjusted to each touchpoint passed by consumers.The flexibility of omnichannel shopping can ensure continuity of channel transitions and make consumers comfortable in shopping, which of course will reduce the risks felt by consumers such as system failures, transaction errors, and low autonomy while shopping [10].
H7: Flexibility significantly affects Perceived Risk in O2O grocery retailers.
Personalization refers to the degree to which customers are able to get personalized attention and tailored service in omnichannel shopping [10].So that, personalization will make consumers believe in omnichannel retailers that they care about individual services, this will certainly increase trust and reduce risk perception from consumers [19].
H8: Personalization significantly affects Perceived Risk in O2O grocery retailers.
From the model produced by Shi et al. [10] which was carried out on multi-industry in China, this study adopts the model, to be applied to the context of O2O grocery retail in Indonesia.Based on PwC [20], China is in the top 4 countries with the highest global omnichannel retail index in 2017.Even though it is a fellow Asian country with a large population, Indonesia is still far behind, even from neighbouring countries.The data confirms that the implementation of omnichannel in Indonesia is still in the introduction stage, which of course has different characteristics from China.That way, the object of research will be a novelty in this study because it describes conditions that are different from previous research, as well as a more specific industry.
The objectives of this study are to prove the influence of connectivity, integration, and consistency to perceived compatibility individually; consistency, flexibility, personalization to perceived risk individually; perceived compatibility to omnichannel shopping intention; and perceived risk to omnichannel shopping intention in O2O grocery retail in Indonesia.

Methods
This is an associative research and Structural Equation Modelling -Partial Least Square (SEM-PLS) is being used as the analysis technique with SmartPLS (v.3.2.9)software to process the research data.The population of this study is O2O grocery customers in Indonesia.Sampling was carried out by purposive method, with questionnaire as data collection technique that is filled by 303 respondents, but the number of respondents who meet the sample criteria (O2O grocery customers) is 251 respondents.The number of the sample is in accordance with what is stated by Hair et al. [21] that recommended sample size for SEM is 200 -400.
The operationalization of variables used in this research is shown in Table 1.Based on Table 1, the questionnaire is developed using five-point Likert scale to collect the respondents' opinion about each variable.Perceived Risk (Y2) [10] Performance Risk (PR1) Financial Risk (PR2) Transaction Risk (PR3) Omnichannel Shopping Intention (Z) [10] Using Omnichannel in the Future (OSI1, OSI2) Omnichannel Recommendation (OSI3, OSI4)

Results and Discussion
The research data comes from 251 respondents as the sample of this research.As shown in Table 2, the gender of the respondents are represented by 60% female and 40% male.While, the most dominant age rage of the respondents is 29 -34 years old as much as 29.5%, followed by age rage 17 -22 years old as much as 28.3%, then age range 23 -28 years old as much as 26.2%, age range 35 -40 years old as much as 10.1%, and more than 40 years old as much as 5.9%.In order to ensure all constructs are eligible to be further processed, validity test using outer loading is implemented.From the validity test, it is known that all the indicators are valid, except Con2, Con5, In1, and Per2.After that, the reliability testing using composite reliability shows that all constructs are reliable.From the results of data processing using SmartPLS, the following results are obtained that can be shown from Table 3 and 4. Meanwhile, the results of the structural model can be seen in Figure 1 below.Based on Table 4, it is known that Connectivity has a significant positive effect on Perceived Compatibility.This is in accordance with research conducted by Shi et al. [10] which states that Connectivity does have a positive influence on Perceived Compatibility.So it can be said that O2O grocery customers in Indonesia attach great importance to connectivity between channels in checking products, reading content, and also account connectivity.This means that the greater the connectivity on each channel owned by industry players, the greater the compatibility perceived by customers.From Figure 1, it can be seen that the Con3 indicator (reading content) has the largest contribution to the Connectivity variable, so the content displayed on the omnichannel platform becomes a very important consideration for customers to feel compatible with grocery O2O retailers.
Furthermore, it is also known from Table 4 that Integration has a significant positive effect on Perceived Compatibility.This is in accordance with research conducted by Shi et al. [10] which says that Integration has a positive influence on Perceived Compatibility.In this study, Integration has the largest coefficient between Connectivity and Consistency in influencing Perceived Compatibility, and from Figure 1 it can also be seen that the contribution of indicators is almost evenly distributed.This shows that product integration and promotion integration between channels are the main things that O2O grocery customers in Indonesia want in order to produce perceived compatibility.
Still from Table 4, Consistency is known to have a significant positive effect on Perceived Compatibility, which is in line with research conducted by Shi et al. [10].This means that the more retailers are able to maintain consistency between their channels, the greater the compatibility perceived by O2O grocery customers in Indonesia.From Figure 1, it can be seen that the Cons3 (product consistency) indicator has the largest contribution to the Consistency variable, so that the similarity of variants and product quality ranges offered by multi-channel retailers are very important considerations for customers to feel compatible with O2O grocery retailers.
Table 4 also reveals that Consistency has no significant effect on Perceived Risk which is contrary to previous research by Shi et al. [10].This is made possible by the different objects and research locations, where in this research which is in the context of O2O grocery which can be said to be a low involvement product because the shopping is done regularly, so that the consistency of retailers in presenting brands, products, and services does not reduce the risk of shopping in omnichannel context.This also applies to the Flexibility variable which has no significant effect on Perceived Risk.This means that in this study Flexibility has not been able to reduce the Perceived Risk perceived by consumers.
However, from Table 4 it can be seen that Personalization is the only variable that has a significant negative effect on Perceived Risk.This is also in accordance with the research conducted by Shi et al. [10].This means that the more retailers are able to personalize their customers through their omnichannel platform, the less risk that O2O grocery customers in Indonesia feel.From Figure 1, it can be seen that the Per3 indicator (personalized webpages) has the largest contribution to the Personalization variable, so that an omnichannel platform that is tailored to individual characters and customer purchase history will be a very important consideration for customers to feel more secure in conducting transactions with retailers.O2O grocery.This is possible because with personalization, customers will feel that retailers know themselves closely so that relationships are established that reduce the risk that customers may experience.
In the end, Table 4 also states that Perceived Compatibility has a significant positive effect on Omnichannel Shopping Intention.These results are in accordance with research conducted by Truong et al. [22] and Silva et al. [23].This means that when customers feel that they are compatible with the retailer's omnichannel platform, the greater the intention of O2O grocery customers in Indonesia to shop omnichannel.From Figure 1, it can be seen that the PC2 (compatible with shopping styles) indicator has the largest contribution to the Perceived Compatibility variable, so that an omnichannel platform that is perceived to be able to meet customer needs for their shopping style will be a very important consideration for customers to display their Omnichannel Shopping Intention at a grocery O2O retailer.This once again confirms the fact that when customers feel familiar with the retailer through the provided omnichannel platform, then the customer will be less reluctant to intend to shop.
However, Perceived Risk is known to have no significant effect on Omnichannel Shopping Intention, which is contrary to research conducted by Han et al. [24]; Shi et al. [10].It is possible that this research was conducted in the middle of the Covid-19 pandemic period, which made customers accustomed to shopping for grocery products in an omnichannel manner, so they did not place too much importance on the risk factors that would be accepted because they had to do this routinely.

Conclusion and Suggestion
The conclusions of this study are Connectivity, Integration, and Consistency are proven to have impact on Perceived Compatibility individually; Personalization is proven to have impact on Perceived Risk, while Consistency and Flexibility have no significant effect on Perceived Risk; and Perceived Compatibility is proven to have an influence to Omnichannel Shopping Intention, while Perceived Risk has no significant effect to Omnichannel Shopping Intention in O2O grocery retail in Indonesia.
Based on the research result, it is recommended that O2O grocery retailers in Indonesia may focus on the development of their omnichannel platforms that accommodate their customers' experience, especially in the context of connectivity, integration, and consistency in order to create make the customers feel compatible with the omnichannel shopping behavior and in the end resulting in the shopping intention.However, this research has limitation, where the technical side of the omnichannel platform which is expected to improve customer experience has not been covered in this study.So that, for further research, it is suggested to study a deeper understanding about this topic by investigating more about the features of the omnichannel platform to provide the omnichannel customer experience mentioned above.The same research model should also be done in other industries to test whether the nonsignificant variables in this research will have the same result in influencing Omnichannel Shopping Intention.

Fig. 1 .
Fig. 1.Structural Model Result Based on Table 3, it can be seen that Perceived Compatibility has an R-Square value of 0.343, meaning that Perceived Compatibility is influenced by the Connectivity, Integration, and Consistency variables by 34.3% and the remaining 65.7% is influenced by other factors outside the study.Furthermore, it can also be seen that the Perceived Risk variable has an R-Square value of 0.150 which means Perceived Risk is influenced by the Consistency, Flexibility, and Personalization variables by 15% and the remaining 85% is influenced by other factors outside this study.And lastly, Omnichannel Shopping Intention has an R-Square value of 0.288 which means that Omnichannel Shopping Intention is influenced by the Perceived Compatibility and Perceived Risk variables by 28.8%, then the remaining 71.2% is influenced by other factors outside the study.Based on Table4, it is known that Connectivity has a significant positive effect on Perceived Compatibility.This is in accordance with research conducted by Shi et al.[10] which states that Connectivity does have a positive influence on Perceived Compatibility.So it can be said that O2O grocery customers in Indonesia attach great importance to connectivity between channels in checking products, reading content, and also account connectivity.This means that the greater the connectivity on each channel owned by industry players, the greater the compatibility perceived by customers.From Figure1, it can be seen that the Con3 indicator (reading content) has the largest contribution to the Connectivity variable, so the content displayed on the omnichannel platform becomes a very important consideration for customers to feel compatible with grocery O2O retailers.Furthermore, it is also known from Table4that Integration has a significant positive effect on Perceived Compatibility.This is in accordance with research conducted by Shi et al.[10] which says that Integration has a positive influence on Perceived Compatibility.In this study, Integration has the largest coefficient between Connectivity and Consistency in influencing Perceived Compatibility, and from Figure1it can also be seen that the contribution of indicators is almost evenly distributed.This shows that product integration

Table 1 .
Operationalization of Variables

Table 2 .
Respondents Profile

Table 3 .
R-Square Value

Table 4 .
Hypothesis Testing Results