Are there any key factors that encourage food delivery applications use during the COVID-19 pandemic in Thailand and the role of HRM?

2.1 Food delivery service

Until recently, people were ordering the food needed for collection in person, even in major cities. Now, new technology has made delivery more convenient for consumers and broadened the range of food available for delivery. At present, the service and industry sectors in Thailand must adjust to support the changing of technology, or what is known as “digital disruption,” which not only causes a change in business methods and production processes but also affects the economy and society of the country through consumer behavioral changes. An example of technology that plays a role in the restaurant business is a food delivery application.

The survey of Kasikorn Research Center [12], found that 63% of the respondents thought that having a food delivery application makes it easier to switch to ordering food and that 88% of the respondents ordered food through the application because of promotions and discounts. The expansion of food delivery and food delivery applications creates the opportunity to generate revenue for those who work in a restaurant chain, which is expected to produce revenue sharing of around $0.8 billion [12]. Food delivery applications had a revenue share of roughly $106 million from the total value of the food delivery business in 2019 (between $1 billion and $1.1 billion), growing by 14% from the previous year and accounting for 8% of the total value of the food business in Thailand in 2019 [12].

The Electronic Transactions Development Agency conducted a food delivery survey and found that 85% of respondents ordered food online, with 51.09% of them being Gen Y. Also, 89% of respondents chose to order the food through the food delivery application, and 61.06% ordered fast food. Gen X spent the most, around $15–$32. Furthermore, the Electronic Transactions Development Agency reported that concern over the situation of the COVID-19 pandemic is one of the critical factors that affect the behavior of consumers in deciding to order food online. The most concerned about the spread of COVID-19 is Gen Y at 40.2% —followed by Gen Z at 30.0%, baby boomers at 28.6%, and Gen X at 26.9%, with a total average of 33.96% [13]. The baby boomer segment was born between 1944 and 1964, Gen X was born between 1965 and 1980, Gen Y was born between 1981 and 1996, and Gen Z was born between 1995 and 2015.

While most restaurants have been turned upside down by the disruptions caused by the Thai government’s reaction to the COVID-19 pandemic, all restaurants have discontinued sit-in dining and now offer takeout through a delivery service. Therefore, the delivery services provide a valuable service by getting food to customers stranded at home while applying social distancing for the safety of their drivers and hungry customers. Food delivery services are among the few local businesses that will survive and thrive during Thailand’s outbreak. The restaurant business has experienced a particularly strong impact of the COVID-19 pandemic. Many restaurants have adopted food delivery to continue doing business. After relying on food delivery and home-cooked meals during the lockdown, diners are now gradually adapting to the “new normal.” At the same time, they remain alert and care for their activities (e.g., eating). The new normal has created a great opportunity for food delivery aggregators because customers are concerned about their health and avoid crowded areas such as shopping malls, restaurants, and cafés. Although these places prepare protection, customers still cannot enjoy the experience as before, so they use a delivery service as their solution. The major food delivery aggregators (such as Grab, LINE MAN, GET, and Foodpanda) served customers with rapid response, offering maximum convenience and transparency with orders (see Table 1). These food delivery aggregators not only enable users to quickly browse menus, prices, and reviews but also allow them to track the progress of their food deliveries in real time. Moreover, online delivery platforms benefit users by enabling them to conveniently explore many local restaurants’ food choices they may not have known existed.

Besides offering local cuisines and street food, restaurants have an excellent opportunity and advantage afforded by aggregators to attract customers through technology. Restaurants can avoid paying a rental fee because they do not need to provide ample space for customers, so this is a competitive point that may save cost and provide a cheaper food cost for the customer. Still, they must face the service and commission fee from food delivery aggregators [8], so restaurant players should focus on balancing costs and customer needs.

Through examining restaurants or cafés and food-related businesses, it might seem like the place for human resource management (HRM) is gradually dwindling. On the contrary, online food businesses need more HR roles than ever. As the online food service and applications continue to boom and expand, particularly during the COVID-19 pandemic, HRM’s role also increases. Although the online food service might be functioning mainly online, employees are still the backbone of the businesses; thus, HR tools are needed. One of the common traits of a successful food delivery service is courteous and reliable staff that work tirelessly to deliver top-notch service to customers. However, with severe competition in the food delivery service sector, hiring the right team and creating the correct HR structure remains one of the biggest challenges. HRM is an integral part of business administration and encompasses important personnel activities to ensure the effective contribution of human capital to the creation and achievement of business strategies [14]. Furthermore, effective HRM is also considered a crucial element of service quality and customer satisfaction [15]. HRM is particularly critical for service businesses that result in a relationship between employee experiences of the organization and customer service experiences [16]. Organizations need to find ways to manage the employment relationship to motivate their employees to provide the support and interaction customers expect. There is a need to align their business strategies and their HRM strategies to pursue long-term competitive advantage [17]. Parry and Tyson [18] revealed that HRM could improve efficiency, service delivery, and market visibility; create a wider customer base; empower managers; and transform HR into a more strategic function. Hence, if companies do not give due emphasis to HR strategies, investment in technologies such as applications or online services may not provide the sought competitive advantage. For service industries whose products involve personal interaction, strategies to achieve competitive advantage inevitably involve employee management.

An emerging sustainable HRM approach might be effective for a firm that adopts a long-term perspective to improve employees’ quality of life to achieve a social sustainability outcome [19]. Moreover, the firm should be aware of creating conditions that facilitate the balance of sustainability behaviors between work and life at home [20]. The sustainable HR approach to performance management as part of the bundle of HRM practices will balance organizational financial goals and provide positive human and social outcomes to gain a competitive advantage. This suggests that sustainable HRM must be carefully tailored to each firm’s situation. Therefore, to obtain greater strength, the sustainable HRM system should enhance the clarity of interpretation of HR practices among employees. Also, business leaders must recognize HR’s strategic role in delivering value to the organization [21].

For companies to survive and be successful in the rapidly changing business world, it is necessary to understand and accept its challenges and adapt business activities to the requirements raised by its characteristics and trends. Identifying risks connected to HR is the first step of such a strategy [22]. One of the biggest challenges for HR managers is how to engage employees effectively during the COVID-19 pandemic. Ahmed et al. [23] found that knowledge sharing could be a possible way to engage employees during the pandemic. Therefore, to maintain the organization’s high performance, employee engagement and knowledge sharing could be the best solution to meet the current HR challenges.

2.3 UTAUT2 (price value, habit, and social influence)

The UTAUT model was developed by Venkatesh et al. (2003) and is widely used by many researchers in many fields. It has been useful to explain about 70% of the variance in behavioral intention and 50% of the variance in technology use [24]. The UTAUT model has four fundamental constructs—performance expectancy, effort expec-tancy, social influence, and facilitating conditions—that influence behavioral intention to use technology. The UTAUT model was initially designed to explain employee technology acceptance and use but not provide context to the consumer. Therefore, Venkatesh et al. [25] proposed a new comprehensive UTAUT model called UTAUT2 to improve analytical ability by paying particular attention to customer use through incorporating the original model with three key constructs: hedonic motivation, price value, and habit (see Fig. 1). Performance expectancy is a concept in line with perceived usefulness. Effort expectancy is also in line with perceived ease of use in TAM, but both expectancies are replaced by convenience to provide a more comprehensive model. Some studies have found that effort expectancy is not significant with regard to the intention to use technology because of the ubiquity of smartphones and the similarity of common features that put the customer at ease in using new applications [26–28]. Lee et al. [26] found that facilitating conditions and hedonic motivation are not key factors that encourage customers to use delivery applications. This may be explained by time usage, which is quite short when customers use delivery applications to order food, so customers still do not enjoy fun and enjoyment from using delivery applications.

OPEN IN VIEWER

Fig. 1

The unified theory of acceptance and use of technology 2 model.

Price is important for the adoption of new technology. Online shopping can offer a lower price to customers because of the lower operating cost of an online business. The monetary benefit from online shopping can be derived from a lower price, promotions and discounts, and a loyalty point system. We followed the idea of Dodds et al. [29] and define price value as a customer’s cognitive trade-off between the perceived benefit of a food delivery application and the monetary cost of using it. The price value is positive when the customer is satisfied that the delivery application’s use is greater than the monetary cost.

Following the idea of Venkatesh et al. [24], we define social influence as the degree to which individuals perceive that important others believe they should use a food delivery application. Social influence as a direct determinant of behavioral intention is represented as the subjective norm in Theory of Reasoned Action (TRA), Technology Acceptance Model 2 (TAM2), Theory of Planned Behavior/Decomposed Theory of Planned Behavior (TPB/DTPB), Combined TAM-TPB (C-TAM-TPB), social factors in Model of PC Utilization (MPCU), and image in Innovation Diffusion Theory (IDT) [24]. The new user who never uses the food delivery application will show the intention to check and gather information on a service or product before deciding by asking, questioning, or searching to increase their perception and understanding. As a person in society, a user is very likely surrounded by friends, family, and colleagues and susceptible to their suggestions. When friends, family, and colleagues of a user start to use a particular food delivery application, the user will be influenced, and the user will adopt it more quickly. As food delivery applications are widely used in society, users strongly consider the opinion of the people they trust since users regard a significant person’s attitude as one of the decision criteria. Likewise, personal satisfaction is given the most priority. Customer satisfaction is often influenced by the quality of the interpersonal interaction between the user and the contact person [27]. In the early stage of adopting the new technology (i.e., a new application), if customers are satisfied with using it for the first time, they will show the intention to use it more often.

Habit is defined as an individual who automatically performs behaviors because of experience using a food delivery application [25]. The empirical findings of Venkatesh et al. [25] about the role of habit in technology use delineate different underlying processes by which habit influences technology use. Habit represents the involuntary behavior to use technology for any activity. Habitual users of food delivery will tend to browse online for food and order it as part of their natural response. As Khalifa and Liu [30] suggested, if customers have the same level of satisfaction with an online store, a customer with high levels of habit will be more likely to repurchase from the same online store than those without such a habit. Such automaticity allows habitual behaviors to be performed easily and quickly [31]. Prior research has reported that habit is a major driver of affect [32]. Affect is defined as “feelings of joy, pleasure, depression, displeasure, or hate associated by an individual with a particular act” [33], an emotional response to the thought of the behavior [32]. Satisfaction is also an effect [30]. HT directly affects satisfaction [32]. Therefore, a customer is likely to be more satisfied with food delivery via an application if they have acquired the habit of application use. We, therefore, offer the following hypotheses:

2.3.2 Application quality

The IS success model was developed by Delone and McLean in 1992 [41]. Many prior studies have attempted to redefine or extend the original IS success model from a different perspective to provide a more comprehensive model. Some researchers found the weakness of the original IS success model [42]. Therefore, Delone and McLean responded and redeveloped the IS success model in 2003 [43]. The updated model consists of six factors: system quality, information quality, service quality, intention to use, user satisfaction, and net benefit, as shown in Fig. 2. The results demonstrated that three characteristics of quality used to evaluate the system affected the intention to use and satisfaction negatively or positively, depending on the net benefit provided. The original IS success model [41] was used by many researchers to evaluate the success of various types of IS—for example, e-commerce [44], the web portal [45], O2O service [7], and mobile catering applications [46]. This study proposes application quality, which combines three main factors of the IS success model: system quality, information quality, and service quality. We define application quality as a multidimensional interface stimulating positive or negative user responses resulting from the interaction between users and an application [47]. Pantano and Priporas’s [48] study found that to encourage customers to adopt mobile commerce, the following features are required: reasonable price, shopping efficiency, transaction security, and service quality. Furthermore, Chi [47] reported that the perceived quality measure is consistent with information quality, service quality, and system quality and that perceived quality positively affects customer satisfaction. The quality of a food delivery application—which is related to visual merchandising, information, the transaction process, and service response—influences customer pleasure and arousal to support quality customer experiences and purchasing decisions. We, therefore, offer the following hypothesis:

OPEN IN VIEWER

Fig. 2

The information systems success model.

H5: Application quality (AQ) positively affects customer satisfaction toward the use of food deli-very applications.

2.3.4 Satisfaction

TAM was proposed by Davis [61]. TAM explains technology adoption with only two factors—per-ceived usefulness and perceived ease of use—and incorporates them with attitude toward technology adoption (see Fig. 3), but in the end, the factors omit the attitude because of a low mediating effect on technology adoption. The attitude expressed by an individual’s positive or negative feelings evaluates technology use [62]. The attitude has both positive and negative expressions, but if consumers have a negative mind-set to use from the beginning, it may cause them to avoid using such technology. Therefore, we propose satisfaction, which shows a positive expression. As an effect of intrinsic motivation, a service provider’s user-friendliness may increase satisfaction with the use of its food delivery application. Customers tend to be satisfied with the use of technology if they perceive it meets their expectations. We define satisfaction as customer judgment that the food delivery application’s product or service provides a pleasurable level of consumption-related fulfillment [63]. The satisfied customers can create a viral effect using technology, especially new technology, if they recommend or share their impressions with others [64]. High customer satisfaction should reduce the operating cost of a company in the future because the company does not need to spend more on acquiring customers [65]. Satisfaction plays a significant role in building and maintaining customer relationships and other vital assets [66]. Tsai et al. [67] demonstrated that satisfied customers are more likely to continue with their intention with a particular organization than dissatisfied customers. Previous empirical studies have supported satisfaction as a positive influence on future intention [68], a key influence on forming a customer’s future intention [69]. To achieve a company’s ultimate goal, the company should devote sincere attention to satisfied customers to increase its competitive advantage. This point is even more critical in e-commerce because the customer does not directly deal with the seller and cannot judge the seller trustworthy [70]. Anderson and Srinivasan [71] reported that customers willing to repurchase from e-retailers believe in assuring an ongoing satisfaction level. We, therefore, offer the following hypothesis:

OPEN IN VIEWER

Fig. 3

The technology acceptance model.

H7: Satisfaction (SAT) positively affects customer intention to use food delivery applications.

This study aims to examine the key factors that facilitate customers’ experiences in terms of satisfaction with the intention to use food delivery applications. The UTAUT2 model was considered theoretically and practically useful to utilize as the theoretical basis in this research. To make it fit food delivery applications, some modifications were necessary. Therefore, this study is designed based on the main indicators of UTAUT2 (social influence, price value, and habit), combined with quality constructs from the IS success model and three variables (convenience, trust, and satisfaction) explicitly proposed to be important in the food delivery application context. The modifications of the model are presented below.

First, performance expectancy, effort expectancy, hedonic motivation, and facilitating conditions were excluded from the model, although they have been proven in prior studies to be significant [25]. Considering an example of dining activity, performance expectancy is not required. Likewise, given the overlap between performance expectancy and perceived usefulness, it has been found that the impact of performance expectancy toward intention might be uncertain, as findings for the impact of performance expectancy on adoption have been mixed [9, 74]. Since mobile applications are omnipresent nowadays and due to the intuitive use of food delivery applications, effort expectancy does not exist. Previous studies have also proved the nonsignificance of effort expectancy [9, 27]. As consumers use food delivery applications as a beneficial service rather than to have fun or entertain, hedonic motivation was also not considered in the model. Also, prior studies have failed to prove the role of hedonic motivation [75, 76]. Experienced users are not concerned about facilitating conditions because they have gained experience and know how to use food delivery applications. Also, no huge effort is needed to use these applications as a navigator to guide the user. Moreover, these applications are designed to fully support users. Recent studies also support the nonsignificant impact of facilitating conditions [27, 73].

Second, application quality and consumer satisfaction levels were included in the research model because perceived quality and satisfaction are at the core of motivating consumers to adopt food delivery applications. Previous research suggests that consumer perceptions of quality determine technology acceptance [46, 47]. The importance of customer satisfaction does not discriminate, as it can make or break a business’s success. The more satisfied users are, the more likely they are to spend more and recommend a food delivery application to friends and family members. The studies of Alalwan [9] and Wang et al. [46] prove that satisfaction facilitates and increases customer intention.

Third, with the advent of food delivery services, the standards of convenience have risen even further. To succeed, restaurant operators will have to continuously up the standard and make it more convenient for consumers to get fresh food delivered to their doorsteps. More importantly, there is a consensus in the literature on food delivery services that convenience plays an important role [8]. Trust is another critical factor in consumers’ online purchasing decisions. The lack of trust is a fundamental reason why consumers do not purchase a product from a site. Many prior studies on mobile commerce have noted that the greater consumers’ trust, the more likely they are to engage their intention [74, 77]. The conceptual model, including the proposed path hypothesis, is shown in Fig. 5.

OPEN IN VIEWER

Fig. 5

The research model.

This empirical research is based on a nonprobabilistic sampling method. The scale items used to measure the construct were selected after an intensive and careful review of relevant literature. We chose questionnaires that were developed based on an energetic body, a well-documented scale, and a match with the current research context. The items of price value were adapted from Venkatesh et al. [25], convenience was borrowed from Liu et al. [78], application quality was proposed by Mutaz et al. [45], and subsequent conceptualization of information quality by Lee et al. [26]. Trust was developed from Xiao et al. [6]; social influence, habit, and intention to use were derived by Lee et al. [26]; and satisfaction was based on the instruments proposed by Ashfaq et al. [79]. Specifically, the data set was collected in Thailand. The questionnaire was first translated into Thai by the authors, and it was tested as a pilot study among 20 respondents who had experienced using food delivery. Based on their feedback, we revised some items to improve clarity and comprehensibility. Then the authors translated them back into English to ensure consistency. Moreover, Cronbach’s alpha was used to check the reliability, and a pilot study revealed that all the constructs captured more than 0.7, which proves that all the constructs are reliable [80]. The final scales and items are presented in Appendix 1.

3.3 Sampling and data collection

Recently, there was a boom in the delivery service business. This was due to various factors that allowed for rapid expansion, whether it is the rapid growth of the e-commerce market, B2C online customer behavior, or technology trends that makes applications smarter and better respond to user needs. The food industry was also affected by these reasons. The delivery service business started to provide more food deliveries to match consumer demands. The food delivery service is growing faster than expected. In addition to delivery, the platform can do much more, creating a situation that benefits all parties. The delivery service providers’ new strategy has offered aggressive service tightly related to the customer’s lifestyle to engage users continuously. At present, many factors (e.g., COVID-19) may have caused the “new normal” in customer consumption and customer behavior. Among these changes, this study focuses on how food delivery leads to satisfaction toward the intention to use food delivery applications. The target respondents are smartphone users who have used a food delivery application. Hence, this study seeks to understand customer attitude and examines the extent to which satisfaction with food delivery applications affects customer intention to use these applications. We developed exploratory and confirmatory research with a quantitative approach. The survey instrument to test the hypothesis and propose the research model was developed from prior studies. The relevant measures and scales were adapted and slightly modified from studies to prove the research questions’ consistency and validity. The modification was made to fit the specific context of food delivery applications. Each item was evaluated using a five-point Likert scale, from strongly disagree = 1 to strongly agree = 5. This research model contains eight constructs; each construct was measured with multiple items.

Data collection took place from April 1–30, 2020, via an online survey administered by Google Forms. Invitation links for the survey were sent to 304 prospective respondents who wanted to participate. Fifty-four respondents were deleted because they returned incomplete surveys. Therefore, surveys from 250 eligible respondents were collected. Tables 2 and 3 summarize the demographic and behavioral characteristics of the participants. It was found that the sample had an almost equal number of males and females. Most of the participants (32.0%) were 31–35 years old. Regarding education level, the majority (60.0%) had bachelor’s degrees; personal monthly income data showed 48.8% were earning between $500 and $1,000; and 43.2% were office workers. In terms of their usage behavior, most of the respondents (54.8%) use Grab delivery service; 43.6% use it two to three times per week, and 39.6% reported an average spending of $3.05–$3.10 per time.

4.1 Measurement model results: Confirmatory factor analysis model fitness was tested initially in the CFA, followed by evaluating the reliability and validity constructs

To achieve an acceptable model fit and confirm that this model is appropriate to predict adoption of the food delivery application, it was necessary to eliminate 3 of the original 27 indicators. The indices’ results, along with their recommended values of model fit, are shown in Table 4. We conclude that the goodness-of-fit indices met the recommended levels, suggesting that the research model provides a good fit for the data [81].

To validate the survey instrument, we examined the convergent and discriminant validity. Convergent validity is evaluated by measuring the standardized path loadings, composite reliability (CR), and average variance extracted to prove reliability and validity. Table 5 indicates all reflective item constructs. The item loadings reported ranged from 0.760 to 0.912, and according to Hair et al. (2010), factor loading estimates should be higher than 0.5 and ideally 0.7 or higher. In the present study, the value of CR reported ranged from 0.835 to 0.928; CR is usually satisfied by a recommended value of 0.7 (for an exploratory study, 0.6 could be accepted [81]. The value of average variance extracted reported ranged from 0.684 to 0.812; the recommended value of AVE should exceed 0.5 [82]. This finding indicates that the items meet the requirement of convergent validity, and all indicators are closely related to the given constructs. Discriminant validity is valid when the square root of each construct’s AVE is larger than its correlations with other constructs [83], as shown in Table 6. The results show that discriminant validity is satisfactory for all constructs, implying that the model meets the criteria to support discriminant validity.

The relationships between and among factors in terms of the combination of direct and indirect effects are referred to as the total effects. Table 7 shows the overall total effect of each prediction factor on each endogenous factor. For example, the standardized total effect of application quality on intention is 0.473, which means that when application quality goes up by 1 standard deviation, the intention will go up by 0.473 standard deviations. Conversely, the total effect on intention from habit is –0.146, which means that an increase of habit of 1 standard deviation would predict a change in intention by –0.146 standard deviation.

To analyze the structural model, as shown in Fig. 5, the model, path coefficients, and R² for the endogenous construct were tested. The x²/df for the structural model was 1.613 and by chi-square was 371.20 with Degree of Freedom (DF) being 230 (p = 0.000). The GFI = 0.892, adjusted GFI = 0.859, NFI = 0.931, CFI = 0.972, and RMSEA = 0.050. These fit measures are comparable to those obtained with the CFA and show a good fit measure. The path coefficients are found at both significant and nonsignificant paths, the significant paths found at the p = 0.001, p = 0.05, and p = 0.1 levels, with price value to satisfaction path at p = 0.638, the social influence to satisfaction path at p = 0.096, the habit to satisfaction path at p = 0.092, the trust to satisfaction path at p = 0.024, the convenience to satisfaction path at p = 0.036, the application quality to satisfaction path at p = 0.000, and the satisfaction to intention path at p = 0.000. The price value to satisfaction path coefficient at β= 0.04, the social influence to satisfaction of β= 0.17, the habit to satisfaction of β= –0.17, the trust to satisfaction of β= 0.20, the convenience to satisfaction of β= 0.16, the application quality to satisfaction of β= 0.56 (relatively stronger than the other paths to satisfaction), and the satisfaction to intention of β= 0.84.

The R² measure is used to explain the proportion of the dependent variable by influencing variables. For the overall R² for satisfaction and intention, the dependent variables are 0.77 and 0.71, respectively, which is considered strong (see Table 5). The model testing in Fig. 5. shows that one of the seven hypotheses, price value (H1), is nonsignificant; two of the seven hypotheses, social influence (H2) and habit (H3), are supported at a significance level of p < 0.10; two of the seven hypotheses, trust (H4) and convenience (H5), are supported at a significance level of p < 0.05; and two of the seven hypotheses, application quality (H6) and satisfaction (H7), are supported at a significance level of p < 0.001. To summarize, the model provided support for six hypotheses (H2–H7) in their predicted directions.

5.1 Theoretical implications

The primary purpose of this study was to understand the key drivers of food delivery applications. We propose that food delivery applications are driven by customer satisfaction—which, in turn, is driven by six relational factors: price value, social influence, habit, trust, convenience, and application quality. Empirical support was found for six of the seven research hypotheses (H2–H7), as shown in Fig. 6.

OPEN IN VIEWER

Fig. 6

Results of the structural modeling analysis.

Somewhat surprisingly, of the main predictors of satisfaction toward adoption—namely, price value (H1), which is one of the main predictions of a fundamental theory (UTAUT2)—price value was not validated as a significant predictor of satisfaction toward adoption, which shows the opposite result from the previous study [25]. Notably, this empirical finding supports the notion that food delivery application users have discovered there are no additional price benefits between placing a food order with/without a delivery application [26], and financial concern seems to play an insignificant role because of decreasing cost (e.g., delivery fee, commission fee) [84]. The invalidated relationship between price value and satisfaction toward adoption exhibits that, given today’s food delivery application toward pleasant feelings, the user’s perceptions regarding commission fees, discounts, or coupons using food delivery applications do not represent a strong indicator anymore. This may imply that the competition among food delivery aggregators has intensified, so they already provide a reasonable price and always launch their promotions or special events to encourage users to use food delivery services. Therefore, prices do not help users decide whether to opt-in or not [76].

Also, the results provide empirical clarification that the paths from the two factors social influence (H2) and trust (H4) are mediating effects of satisfaction toward adoption. The strength of social influence is consistent with the original studies of UTAUT on information technology [24], late IS theory [25], and other recent conceptualizations of IS adoption [26, 85]: users will be more willing to use technology if they gain a favorable opinion from friends, family, or referents. However, some studies in mobile commerce [27] and airline e-commerce [75] have found a nonsignificant influence of social influence on intention to use technology. Moreover, the findings on trust agree with prior studies in e-commerce [60, 86] that trust has a long-term impact on future relationships and intention through satisfaction, and it can be concluded that a greater trust of the user results in greater intention to use any service.

The empirically proven factor habit (H3), as a significant factor negatively determining the satisfaction to adopt food delivery applications (β= –0.17; p < 0.10), implies that it is an important aspect considered by users in formulating their satisfaction to use or reject food delivery applications. In addition, the negative sign indicates that users are less likely to be satisfied and discouraged to use food delivery applications, with a higher degree of expectation of recovery from COVID-19. The spread of COVID-19 has forced consumers to cope with changing their eating habits. Being stuck at home, consumers tend to eat more to cope with stress or boredom, and right now, people find an abundance of both [87], which may be the cause of a negative relationship. Moreover, current eating habits require social distancing, which may not form actual customer intention, so eating habits will be in constant evolution throughout the upcoming year as consumers decide which habits to keep and which to return to the “old normal.” As McKinsey’s [49] survey found, consumers plan to resume some of their out-of-home activities soon, and dining at a restaurant or bar is on the list. This proves that if safety measures are in place and the development of a COVID-19 vaccine is confirmed, consumers will start the shift to their “normal” out-of-home activities, which directly affects the intention to use food delivery.

Convenience (H5) is a factor that plays a crucial role as a direct driver of satisfaction, provided the most convenient access to save time and effort in using applications lead to a stronger feeling of satisfaction. This critical role of convenience is supported by studies on customer experience [40] and usage transfer behavior [7]. This result reflects the convenience-oriented user’s decisive perception of a food delivery application matching their pleasant feeling.

Our results show that satisfaction is usually derived from application quality (H6). This result is consistent with studies that have demonstrated the important role of application quality in the context of m-commerce [47] and e-service [88]. Online customer satisfaction forms during the interaction between customers and components of the application and service [88]. Continued updates in application quality and service provision that meet customer needs and expectations will improve customer shopping experiences toward intention to use [47].

This study supports the five factors that perform the role of mediator satisfaction and exhibit a high explanatory power, which explains 77% of the variance. Finally, the results reveal that satisfaction (H7) is the critical dominant factor toward intention to use food delivery applications, with a high explanatory power of 71% of the variance. This result coincides with IS studies that have confirmed that satisfaction determines the user’s continuance intention to use [9, 89] and that satisfaction has a positive effect on the intention to recommend [88]. Our result implies that customers who adopt food delivery applications are more likely to evaluate such a claim as satisfied and raise their intention to use.

In addition to the theoretical implications, the results of this study also provide implications for practice. First is the validation of application quality as the strongest predictor of satisfaction toward the intention to use. As a result, food delivery aggregators or retailers should develop high-quality applications that provide a seamless experience between customers and the applications, which would provide future benefits [45]. Moreover, such an application should have interactive features that respond in a way that feels natural to the user, the presentation of products should be improved practically, and the application should be implemented to attract customer interest to stimulate and place an order [36, 89]. Besides, extra efforts should be made to improve the quality of the system (e.g., features, the payment system, real-time tracking), information (e.g., delivery area, event, certification), and service (e.g., delivery, in-app chat) for the fulfillment of the customer aspect and their shopping experience.

Second, both social influence and trust show a mediating effect on satisfaction toward intention to use food delivery applications. Therefore, food delivery aggregators should deploy advertising via social media and focus on extroverted users—who are likely to share an experience from their previous use, which will encourage potential users to use the applications. These methods will help general users become aware of the benefits of using food delivery applications [85]. Moreover, the marketing strategy should focus on reputation building to gain positive feedback from users and should promote and encourage user intentions and trust [27, 77]. The information quality and security assurance provided by these applications will help to improve trustworthiness. Proper security and privacy protection should be implemented to enhance and maintain perceived trust from users [77]. Offering useful and complete advice will eventually result in a higher intention to adopt food delivery applications.

Third, this study shows that habit is a significant factor hindering satisfaction toward the intention to use. Therefore, food delivery providers or restaurant players should wait to observe customer behavior after COVID-19 recovery to propose the proper consumer retention strategy and maintain customer loyalty. Moreover, according to a research study from University College London, forming a habit takes an average of 66 days [90], and Thai customers have already surpassed that mark. This means that new habits and practices have formed already, and consumers will soon begin to blend their new practices with old routines [91].

Fourth, the role of convenience encourages customer satisfaction and intention. Hsu et al. [39] and Jiang et al. [36] suggested that convenience provides additional improvement and minimize service time, search efforts, conditions, and energy spent to meet customer expectations. Therefore, business operators should provide the easiest and fastest service and effort to aid customers’ perception of convenience and enhance satisfaction. Food delivery aggregators offer greater convenience than restaurants; hence, they would benefit more and earn a competitive advantage. As food delivery often involves a payment method and a service person during food shopping and delivery, offering users opportunities to choose the form of payment and the service person may help them derive more satisfaction.

Last, it is important to address satisfaction because it plays a vital role in increasing customer intention and competitive advantage. Customer satisfaction is paramount in sustaining a business in a highly competitive market [40]. Also, food delivery aggregators and retailers should maximize customer expectations and needs during customers’ use of food delivery applications. This will strengthen the relationship with customers [79, 92]. Therefore, food delivery aggregators and retailers should fulfill customers’ needs and understand the ingredients to achieve customer satisfaction.

5.3 Managerial contributions

This study’s findings can be interpreted to provide specific practical/industry solutions in order for the food industry, consumers, and academia to appropriate value from their experiences. The food industry can utilize the findings in their food delivery services.

In this regard, when setting a service fee or commission fee, a business must make sure that the fees set are reasonable and appropriate with customers’ living expenses while bearing in mind the fees its competitors charge. Special treatments are also necessary to keep customers loyal to a business and attract other customers.

Marketers should also provide more value to their customers, such as helpful tips that benefit them, and then their followers will engage and share with others in return. Building an engaged excited following requires tapping into emotions and paying attention to customer needs. Followers who are engaged in a business will more likely share its content with others, thereby expanding its influence. Creating high-quality content and a good social media campaign can help to drive brand awareness, increasing customer retention and/or boosting sales and reputation.

Food delivery practitioners must be aware and make sure their customers are safe when they shop with them. Clearly showing contact information and being always available through responding quickly puts customers at ease, knowing that when they encounter problems, they will be able to get in touch with a professional customer service.

Customers who wish to engage in food delivery services want to experience convenience. Food delivery practitioners should ensure that they optimize the customer menu for delivery and that customers enjoy a fast delivery time. The food delivery applications should be easy to access to help customers find popular restaurants to order some delicious food and easy to use to make an order. Multiple options for payment, packaging, and so forth should be provided, making customers feel more comfortable purchasing from the service. Offering customers opportunities to choose ingredients simply and effectively also provides an advantage. The key to delivery is that it should be fast, smooth, and satisfactory for every customer.

Developing and improving food delivery applications could enhance the customer’s experience of ordering food by focusing more on the quality of food delivery applications. Food delivery applications should be designed in such a way that customers are afforded interactivities and responsiveness with the application. Therefore, a minimalistic design approach should be adopted to create a simple and highly usable application. With the massive number of food delivery users, the existing technical support must have the ability to support concurrent users and do so very quickly without crashing. The easy navigation design, advanced search, consistency, and information provided will all improve the application’s usability to a great extent. It would be best to offer live chat assistance within the application through which customers can live chat with the support team whenever they need to. Displaying too much content should be avoided; in contrast, more and more images should be provided to make the application more interactive. Moreover, providing the ability to sign up via social media accounts and enabling customers to share content from the application is great for extending services to other customers.

Besides, as mentioned above, in terms of improving service effectiveness, customer satisfaction, and higher business performance, food delivery providers and restaurant operators should understand and cultivate good HRM practices—as food businesses are associated with human capital and need to carefully manage, particularly during the COVID-19 pandemic. Tensay and Singh’s [93] study revealed that HRM systems, employee engagement, and organizational performance have positive and significant relationships with each other. Thus, implementing proper HR practices helps to enhance the effect of performance on both the individual and the organization. Therefore, attention should be paid to HRM practices—adapting and implementing them through investing a significant amount of time, thereby improving business performance and enhancing employee productivity. As a result, HRM will help reduce business costs, achieve business goals, and build an engaged workforce that is satisfied, motivated, and committed to providing the expected service. Moreover, providing a mutual benefit perspective between the organization and individual employees must involve being resilient to cope with work challenges imposed by the organizational-level bundles of sustainable HRM practices.