Introduction

COVID-19 has started to gain global attention since January 2020. The World Health Organization (WHO) officially declared the coronavirus (COVID-19) outbreak to be a global pandemic on March 11th, 2020 (WHO 2020). Many countries create various regulations to stop the spread, such as implementing a strict quarantine policy, work-from-home (WFH), lockdowns, and activities shut down that require human gathering and interactions—including colleges, schools, malls, offices, airports, and many others. This results in most individuals adopting technology to meet their daily needs, such as communication, studying, working, shopping, among others. (De’ et al. 2020). One significant example is the increasing number of Internet users during this outbreak that leads to a significant surge of digital payment new users. WHO and the government encourage the use of digital payments. Moreover, offline stores that have reopened are also implementing cashless payments to avoid contact, (Huang 2020) because the COVID-19 virus is easily spread and transmitted through money. The situation underlies the digital payment users’ surge. Bank of Indonesia recorded an increase in digital payment transactions since the implementation of the social distancing, reaching 64.48% and the transaction volume grew by 37.35% (Nabila 2020). Based on MarkPlus survey, digital payment platforms in Indonesia that experienced a rapid increase during the pandemic are; ShopeePay, OVO, GoPay, and Dana (Santia 2020). New OVO users, for example, recorded an increase of 267%. Transaction growth at DANA from January to mid-May reached 50%. As for GoPay, a GoJek's e-wallet recorded an increase in transactions up to 103% (Astutik and Hastuti 2020). GoPay has been used by more than 500,000 business partners, 97% of which are SMEs (Jatmoko 2020), and transactions for groceries at Gojek via GoPay recorded an increase of up to 500% (Burhan 2020). Likewise, transactions on ShopeePay during the pandemic increased six times and up to eight times in Jakarta (Sarifah 2020). This study focusing on Indonesian older generation since Indonesia is the sixth largest Internet users in the world and shaping up to be the next Southeast Asia’s digital payments battleground with the high growth of digital payment users, competitiveness, and the influx of foreign payment services and investment, i.e., Alipay and Paypal. Quoting the report from Google, Temasek and Bain Company in the e-Conomy SEA 2020 report, Indonesia had become the country with the highest digital economic transaction value in Southeast Asia, amounting to US $44 billion (Google 2020). This supported by the contribution to various digital services in Indonesia which are the top leaders in Asia such as Tokopedia, Gojek, and Ruang Guru (Google 2020).

Based on data from Statisika (2000), there is also an increase in digital payment users in Indonesia in 2020, including the older adults. For instance, 92.74% of respondents aged between 35 and 44 stated that they had used digital payments, 89.37% for those aged between 45 and 54 and 81.25% for those over 55 years old. This phenomenon is the focus of our research. We want to analyze whether new digital payment users who are X generation and baby boomers will continue to use digital payments after the pandemic. It is uncertain when this pandemic will end. With the long periods of forced use of technology for older adults, it allows for a significant change in their attitudes toward technology acceptance (Magsamen-Conrad et al. 2020). Analyzing continuance intention (CI) toward digital payments of the older generation in a pandemic situation is important, considering that it can be a new opportunity for digital payment providers to expand their target market during and after this pandemic, but there is still a lack of study on this. They can create a more appropriate services and marketing strategy, particularly the stereotype and some previous studies show that older adults are ignorance to technology.

We proposed the extended Unified Theory of Acceptance and Use of Technology (UTAUT2) (Venkatesh et al. 2012) to answer this question. Theoretically and practically, TAM and UTAUT are the main conceptual framework and they are the most used by researchers in analyzing the acceptance of information systems and technology. From previous research comparing various technology acceptance theories (de Abrahão et al. 2016; Al-Saedi et al. 2020; Hamrul et al. 2013; Junadi 2015; Patil et al. 2017), it can be accepted that the UTAUT method is the best method used in this case study, especially in COVID-19 pandemic situation where many people forced to use technology. Previous studies on digital payments in various countries such as Australia (Gao and Waechter 2017), Brazil (de Abrahão et al. 2016), Nigeria (Gholami et al. 2010), Germany (Hohenberg and Rufera 2004), India (Liébana-Cabanillas et al. 2020; Patil et al. 2017; Sahu and Singh 2018), China (Li and Zhang 2012; Zhou 2011), and Indonesia (Junadi 2015) discuss digital payment adoption and intention to use. Nevertheless, there are still limited studies on CI in digital payments. Previous research on this topic (Maureen Nelloh et al. 2019) only measures the trust factor, whereas in our opinion, the trust factor alone is not enough to measure continuity intention toward digital payments in the midst of a pandemic situation. We argue that UTAUT2 that was used by Junadi (2015), Venkatesh et al. (2012), inertia used by Wang et al. (2019) and satisfaction used by Huang and Cheng (2012), Wang et al. (2019) need to be added to analyze CI in digital payments.

This study offers several implications that could help practitioners, researchers, and digital payment providers evaluate technology acceptance and CI in the older age range who are increasingly using digital payments amid this COVID-19 pandemic. This study contributes to the existing literature on digital payment services by investigating the relationship between several key factors (i.e., UTAUT2, satisfaction, and inertia) and CI toward digital payment services after the outbreak has ended. The findings of the study will be beneficial for practitioners as well. There is a lack of knowledge on the factors influencing satisfaction toward Indonesian customers’ continuance intention to use digital payment, especially the older generation. This will help digital payment providers by showing the main antecedent factors to predict the CI of the older generation.

Theoretical background and hypothesis development

Baby boomer and X generation

There are seven classification of generational living, namely the greatest generation, silent generation, baby boomers, X generation, Millennials, Z generation, and Generation Alpha (Childs et al. 2015; Rudolph et al. 2020). These are generational births defined as a cohort of people born within a particular period of time (Sandeen 2009). From the seven generation, it has birth range from generation to other generation (Koksal 2019). The Greatest generation was born from 1901–1927, Silent generation was born from 1928–1945, baby boomers were from 1946–1964, X generation was from 1965–1980, Millennials were from 1981–1997, Z generation was from 1998–2010, and Generation Alpha is 2011–2025 (Bejtkovsky 2016). From a marketing point of view, this generational living has unique characteristics in social and in buying behavior (Berkup 2014). Baby boomers, for instance, tend to buy products in store. Representative stores as a need in terms of branding, strategic place, or distribution area are their concern. Interactive in service quality is one of greatest values, and they always find products with a good value for money (Roberts and Manolis 2000). Meanwhile, X generation tends to research a product or service they want before taking decision. They buy high-quality products and great service. They know about technology but old school tech such as desktop or notebook computer compared to Millennials and Z generation who prefer to use smartphone. Besides, X generation likes to shop in physical stores (Lissitsa and Kol 2016).

The unique characteristic of Millennials is that they tend to research products or services on the Internet. They like only marketing channels that are logical to them such as word of mouth, content marketing and social Selling. They require honesty from a brand such as the ability of a brand to deliver great values because that means that company has a good management. Price is also one of the elements that benchmark to every company they need (Berkup 2014). The character of Z generation is that they tend to despise brand distrust. They like an advertising that shows the real of all elements. Social media has much affected their buying behavior. Besides, this generation has an awareness to manage the fund for buying and saving. They do research on products only with their smartphone, and they can spend more than one transaction (Priporas et al. 2017). Meanwhile, Alpha Generation’s tendency in shopping is influenced by family purchase. Every alpha has an e-commerce account. They are not surprised to artificial intelligence services. They learn digital technology faster, and they always need fresh products rather than old products even though the older products might have better values (Rani Thomas et al. 2020). Baby boomers were 56–74 years old in 2020; they were born between 1946 and 1964 (Lim et al. 2011; Littrell et al. 2005). X generation was born between 1965 and 1979; it integrates those between 41 and 55 years old in 2020 (Berraies et al. 2017). They spend a lot, exhibit a brand loyalty, are advertising literate, financial aware, and respond to changing trends (Guido et al. 2020; Lim et al. 2011). However, the baby boomers and X generation are the forgotten generations. They have not been receiving enough attention from the marketers who have been more aggressively targeting the younger generations, particularly in digital products or services. Contrary to the stereotypes that elderly are difficult to adapt to new technology, several recent studies show that baby boomers and X generation are quite experienced in Internet usage, they do not feel the technology anxiety (Magsamen-Conrad and Dillon 2020; Niemelä-Nyrhinen 2007), have a positive attitude toward online shopping intention (Lim et al. 2011) and are interested in using mobile banking (Shams et al. 2020). Hence, baby boomers and X generation may become new opportunities for Internet marketers.

Unified theory of acceptance and use of technology (UTAUT)

UTAUT is a theory that aims to explain user intention in using an information system (IS) (Persada et al. 2019). It can be measured by how market accepts a new technology (Andwika and Witjaksono 2020). It combines eight models of technology acceptance; theory of reasoned action (TRA), the technology acceptance model (TAM), the motivational model (MM), the theory of planned behavior (TPB), a model combining TAM and TPB, the model of PC utilization, the innovation diffusion theory (IDT), and the social cognitive theory (SCT). UTAUT consists of four constructs: performance expectancy (PE), effort expectancy (EE), social influence (SI), and facilitating condition (FC). Given that UTAUT explains for as much as 70% of the variance in intention (Venkatesh et al. 2003), it can be aimed as a tool that assesses the likelihood of success in introducing a new technology targeted at a population of users who may be less interested in adopting and using a new technology, as in this study. In this study, we used UTAUT2, an extending version of UTAUT, which incorporates three constructs; hedonic motivation (HM), price value (PV), and habit (Viswanath et al. 2012). The UTAUT method is able to measure as much as 70.7% of the aspects that can be used to improve technology acceptance, and TAM is only able to measure as much as 62, 1% (Hamrul et al. 2013). Venkatesh et al. (2012) shows that there is an increase in the value of variance in behavioral intentions in the UTAUT2 model compared to the UTAUT model, as well as an increase in the variant value of technology use in UTAUT2. Thus, it can be accepted that the UTAUT2 model can explain behavioral intentions and use of technology better than the UTAUT model, and it can explain this phenomenon in a pandemic situation since there are new habit and effort expectancy constructs. The theory of UTAUT2 can fit into this study because it can understand user behavior toward digital technology (Persada et al. 2019), especially in classification of birth generation like baby boomers and X generation as a potential market in digital payments industries (Magsamen-Conrad et al. 2015; Verster et al. 2018). However, in this study, we did not include the HM construct. We examined new digital payment users in a pandemic situation, where many of them started using digital payments because the needs to avoid contact. Thus, HM became irrelevant. Prior study (Alalwan et al. 2017; Venkatesh et al. 2012) shows that HM factors explain customers’ hedonic motives in using technology such as entertainment, pleasure, joy, and enjoyment.

Previous study found that UTAUT2 construct affects users’ satisfaction (US) and CI in mobile food ordering apps (Alalwan 2020). However, there are still limited studies explaining UTAUT2 to US and CI in digital payments. Thus, this study will develop and expand previous research by analyzing UTAUT2 construct effect on US in the use of digital payments. The following hypotheses are proposed.

Performance expectancy (PE)

PE illustrates the consumer’s belief that using certain innovations will help improve their job performance (Venkatesh et al. 2012). Digital payment provides convenience for customers with a flexible time and place to access it, especially in this pandemic situation (De’ et al. 2020; Patil et al. 2017; Sahu and Singh 2018). Accordingly, it can be proposed that as long as customers perceive using digital payment can help them improve their performance and productivity, they will be satisfied with their experience using digital payment. Prior study indicates that PE positively affects user satisfaction (Alalwan 2020). Thus, this study assumed the following hypotheses.

H1

PE has a positive effect on US.

Effort expectancy (EE)

EE is the perception that the use of technology to facilitate the work can be easily learned (Venkatesh et al. 2012). Accordingly, it can be proposed that as long as customers perceive using digital payment is effortless, they will be satisfied with their experience using digital payment. Zhou (2011) supported the impact of EE with users satisfaction in mobile banking area. Alalwan (2020) provided evidence in mobile food ordering apps and further evidence in online learning app (Kaewkitipong et al. 2016). Therefore, the following hypothesis suggests that:

H2

EE has a positive effect on US.

Social influence (SI)

Social influence is “the extent to which an individual perceives that important others believe he or she should apply the new system” (Venkatesh et al. 2003). Since baby boomer and X generation are digital immigrant (Magsamen-Conrad et al. 2015), they are not fully familiar with new technology including digital payment. Therefore, to get to know a new technology, they are usually influenced by other people they trust and the people around them can also influence their assessment of their experience (satisfied or dissatisfied) in using digital payments (Alalwan et al. 2017; Hsiao et al. 2016) and they will be satisfied when they are capable to use what was recommended by their colleagues. Thus, the following hypothesis is proposed.

H3

SI has a positive effect on US.

Facilitating condition (FC)

Facility conditions refer to the ease of access to infrastructure and technical support for technology users (Venkatesh et al. 2012). If customers have adequate facilities to access digital payments, they will feel more comfortable, and it will also affect their satisfaction toward digital payment. The relationship between FC and user satisfaction is also confirmed by several previous studies (Alalwan 2020; Maillet et al. 2015). Thus, the following hypothesis is proposed.

H4

FC has a positive effect on US.

Price value (PV)

PV relates to the financial aspects of using new products and systems. PV is an additional construct in UTAUT2 that differentiates it from UTAUT, making UTAUT2 more relevant to use for the consumer market (Venkatesh et al. 2012). PV is important construct in this study since Indonesians are price sensitive (Hartono et al. 2020). In Indonesia, digital payment offer many discounts, cashbacks, points, coupons, etc. Iyer et al. (2018) and Alalwan (2020) provided evidence about the role of PV and user satisfaction. Accordingly, it can be proposed that as long digital payment provider offer good price value, they will be satisfied with their experience using digital payment Thus, the following hypothesis is proposed.

H5

PV has a positive effect on US.

Habit (H)

Habit can be defined as an act that is spontaneously carried out as an accumulation of learning or adaptation process (Limayem and Cheung 2008). In a long enough time forced to adapt using digital payments (almost 2 years) during this pandemic, a habit will be formed among older adults (Magsamen-Conrad and Dillon 2020). When older adults feel comfortable and accustomed to using digital payments, a habit will form and accordingly lead to a satisfying experience when they are using it (Wang et al. 2013). Thus, the following hypothesis proposes that.

H6

Habit has a positive effect on US.

User satisfaction

Boulding et al. (1993) discusses the concept of users’ satisfaction (US) based on transaction-specific and cumulative perspectives. The transaction-specific perspective refers to the level of US with a transaction in a particular situation. The cumulative perspective indicates that US is developed based on the customer’s overall evaluation of their experience with a particular entity, such as a service delivery system, vendor, or service provider (Wang et al. 2019). User satisfaction is defined as an emotional response based on the user’s overall evaluation of their expectations and experiences after using the service (Oliver 2010). US and CI have a correlation. US can build a continuance intention in consumer. If customers feel satisfied, they will likely continue to use the service/product. In other words, CI is determined by how satisfied consumers are on service channel usage after times of usage (Abu Salim et al. 2020). This construct is important to examine to find out whether older adults are satisfied after using digital payments which will affect their future intentions.

Inertia

Consumer inertia is the habitual choice of commodity or service provider built upon consumer past consumption experience (Solomon et al. 2007). When consumers of mobile service feel comfortable to continue using the service when they are familiar with service content, they will avoid spending any extra time and efforts to learn about new services inhibit switching behavior (Kim and Kang 2016; Lin and Huang 2014; Polites and Karahanna 2012). This comfort makes consumers have emotional attachments and form inertia. Inertia can occur from the satisfaction felt from the previous experiences so that users avoid the risk of moving (Polites and Karahanna 2012). Thus, satisfaction can shaped customers’ inertia. Inertia is highly associated with repeated purchase that defines a strong correlation for consumer to avoid the risk, avoid the variable cost and others like disappointment. They have intention to continue to use or shop at same store or same product if inertia has formed (Wang et al. 2019). Thus, we added inertia to examine since it will likely form continuance intention.

H7

User satisfaction positively affects users’ inertia with regard to digital payment.

Continuance intention

Kim and Kang (2016) defined continuance intention (CI) as users’ long-term use of a specific m-service on a regular basis. IT/m-service users’ CI is influenced by unconscious processes such as inertia, which reflects users’ resistance to alternative services (Amoroso and Lim 2017; Kim and Kang 2016). In this study, we are keen to assess the CI of baby boomers and X generation who begin using digital payments in the midst of the COVID-19 pandemic. We consider users’ CI to be an important indicator of long-term usage of digital payment after the COVID-19 pandemic among the elderly.

UTAUT2 construct used in this study is to examine the acceptance of technology by baby boomer and X generations, whether they find it difficult to use technology or have adapted to technology in the midst of this pandemic. Their acceptance of technology will influence their satisfaction toward it. Polites and Karahanna (2012) indicate that when individuals are satisfied with the ongoing action, they will tend to continue using it. Ashfaq et al. (2020) revealed that satisfaction on e-service is a strong determinant and predictor of users’ CI. The more satisfied the users are, the higher the likelihood that they will continue using the current service/product (Oghuma et al. 2016). Additionally, in the context of technology adoption, prior studies indicate that inertia has a strong positive influence on CI (Wang et al. 2019). Hence, the following hypotheses are proposed.

H8

User satisfaction positively affects users’ CI with regard to digital payment.

H9

Inertia positively affects users’ continuance intentions with regard to digital payment.

Overall, the research hypotheses developed based on the discussion presented above highlight the relationships among UTAUT2 construct, user satisfaction, inertia, CI as presented at Fig. 1.

Fig. 1
figure 1

Conceptual framework

Research methodology

This study uses a semantic differential scale, which is an attitude measurement scale with a 10-point measurement scale (Osgood et al. 1957). The research variables consisted of four variables, namely UTAUT2 which consisted of six indicators and 14 measurement items, users’ satisfaction consisted of four measurement items, inertia consisted of three measurement items, and CI consisted of three measurement item. Several previous studies use to support this research; UTAUT (Venkatesh et al. 2003), UTAUT2 (Venkatesh et al. 2012), and the relationship between users satisfaction, inertia, and CI (Wang et al. 2019). The analytical tool used in this research is structural equation modeling (SEM). The data collection technique used a research questionnaire. The sampling technique used in this study was a purposive sampling technique with the criteria for respondents aged 40–74 years old and only using digital payments during the COVID-19 pandemic. The questionnaire used to obtain the data in this study was distributed to 320 respondents. This number has met the minimum number of respondents, which is five times the estimated parameter, or as many as 285 people.

Variable

Measurement item

Factor loading

AVE

Cronbach’s alpha

References

PE

PE 1: Digital payment provides benefits for my life, especially in this pandemic

0.864

0.816

0.930

de Abrahão et al. (2016), Alalwan (2020), Al-Saedi et al. (2020), Gholami et al. (2010), Junadi (2015), Persada et al. (2019), Pohlmann et al. (2019), Patil et al. (2017), Venkatesh et al. (2003), Venkatesh et al. (2012)

PE 2: I can save transaction time when using digital payments

0.856

PE 3: My transactions are more effective when using digital payments

0.803

EE

EE 1: I can easily use digital payment transactions

0.810

0.798

0.922

EE 2: I clearly understand how to make transactions using digital payments

0.846

EE 3:Digital payment is easy to learn

0.824

SI

SI 1: I use digital payment transactions because my family suggests me

0.810

0.823

0.903

SI 2: I use digital payment transactions because my colleagues / friends suggest me

0.882

FC

FC 1: My cellphone has a feature that can help me make digital payment transactions

0.860

0.862

0.925

FC 2: My Internet connection can help me make digital payment transactions

0.893

PV

PV 1: I get many promos that are profitable when using digital payment transactions

0.923

0.855

PV 1: When it comes to digital payments, the price I pay is more affordable

0.819

0.921

H

H 1: I am accustomed to using digital payments in the transactions I make

0.864

0.867

0.930

H 2: I will have difficulty if I stop using digital payments

0.898

US

US 1: I am satisfied using digital payment transactions

0.868

0.779

0.933

Kang and Lee (2010), Oghuma et al. (2016)

US 2: Digital payment transactions are able to meet my needs

0.712

US 3: Digital payment transactions are able to meet my expectations

0.872

US 4: I’m happy using digital payment

0.794

I

I 1: I never thought of stopping using digital payment transaction services

0.835

0.771

0.910

Kim and Kang (2016), Lin and Huang (2014), Wang et al. (2019)

I 2: I will continue to look for attractive promos provided by digital payment transaction services

0.796

I 3: I will lose if I stop using digital payment transaction services

0.790

CI

CI 1: I will continue to use digital payment transaction services

0.735

0.669

0.858

Hong et al. (2017), Li and Fang (2019), Maureen Nelloh et al. (2019), Oghuma et al. (2016)

CI 2: I will more often use digital payment transaction services

0.733

CI 3: I will continue to use digital payment services if the services are provided by the vendor

0.731

Findings

The number of respondents is 57% dominated by female respondents. This is because women are more frequent in using smartphone and more concerned with discounts, cashbacks, and other benefits obtained from digital payments compared to male respondents. The age of the respondents is dominated by the age of 40–55 years with the number of respondents as much as 72%. This is because the age range of 40–55 years is the productive age for Indonesian people and is easier to adopt technology. Respondents’ occupations are mostly entrepreneur and private employees with a percentage of 37.2% and 32.8%, respectively. This illustrates that those who work in the private sector and entrepreneur are more able to adapt to technological developments by using digital financing. Respondents with incomes above 5 million use more digital payments because their access, facilities, and understanding of digital payments are higher than respondents with incomes below 5 million. The number of respondents who often use transactions via digital payment platforms is 75% (Table 1).

Table 1 Respondent demographics

Measurement model analysis

To test whether a research model has met the appropriate measurement criteria, it is necessary to test the reliability and validity of the research results. Reliability is a test to check the extent to which the measurement of a test remains consistent after being repeated on the subject and in the same conditions (Anastasi and Urbina 1998; Azwar 2011; Neuman 2007). The reliability of the research variables was assessed by calculating the composite reliability of each latent variable. The value of the coefficient recommended composite reliability is above 0.70 (Hair 2010). Composite reliability obtained for each variable construct in this study is above 0.70, thus, it can be said that the measurement in this study is reliable. Average variance extract (AVE) is a test that complements the composite reliability test and reflects the total number of variance indicators capable of representing each latent variable (Hair et al. 2010). The AVE value of each variable must exceed 0.50 (Hair et al. 2010). All of the AVE values ​​obtained in this study were more than 0.50 (Table 2). This shows that the indicators in this study have represented well the latent variables developed.

Table 2 Average variance extracted (AVE), variance extract (VE), and composite reliability (CR)

Validity is the accuracy of interpretations made from the results of measurements in a study (Gronlund and Linn 1990). The validity of the measurement model is determined by conducting a convergent validity test and discriminant validity. Convergent validity test of the measurement model is estimated by calculating the composite reliability of each construct and AVE. The coefficient value of the composite reliability variable must be above 0.70, while the AVE value of the entire variable must exceed the 0.50 value limit (Fornell and Larcker 1981). Table 2 shows all the CR and AVE values ​​of the measurement models in this study are above the recommended values ​​so that the measurement models in this study pass the convergent validity test. Discriminant validity was assessed by comparing the AVE root values ​​with the correlation square between constructs (Fornell and Larcker 1981). The results (Table 3) show that the correlation square between the constructs is lower than the AVE root so that the measurements in the research model pass the discriminant validity test.

Table 3 Discriminant validity of measured items

The high level of correlation between price value and habit, price value with facilitating conditions and price value with effort expectancy is because baby boomers are an age group that has a high level of price sensitivity. Baby boomers are more likely to choose a product that is cheap or more profitable economically than other factors (Worsley et al. 2010). Thus, the correlation between PV and H, FC and EE is high. When the boomers are feel the benefits of digital payment economically, they respond very well to the assessment of EE, PC and H.

Goodness of fit

The Goodness of fit test in SEM analysis is carried out to ensure that the research model is a fit model and can be used in other research works. The model suitability test can be seen by comparing the results of the chi-square, probability, CMIN/DF, GFI, AGFI, TLI, CFI, and RMSEA analysis results with cutoff value. Then, this goodness of fit test can be accepted or the model is categorized as fit if at least 5 of these criteria are met (Hair et al. 2010). The results of goodness of fit test can be seen in Table 4 as follows: After analyzing the data using SEM with AMOS 23.0 software, the results of goodness of fit are obtained as follows:Validity is the accuracy of interpretations made from the results of measurements in a study (Gronlund and Linn 1990). The validity of the measurement model is determined by conducting a convergent validity test and discriminant validity. Convergent validity test of the measurement model is estimated by calculating the composite reliability of each construct and AVE. The coefficient value of the composite reliability variable must be above 0.70, while the AVE value of the entire variable must exceed the 0.50 value limit (Fornell and Larcker 1981). Table 2 shows all the CR and AVE values ​​of the measurement models in this study are above the recommended values ​​so that the measurement models in this study pass the convergent validity test. Discriminant validity was assessed by comparing the AVE root values ​​with the correlation square between constructs (Fornell and Larcker 1981). The results (Table 3) show that the correlation square between the constructs is lower than the AVE root so that the measurements in the research model pass the discriminant validity test.

Table 4 Goodness of fit

Based on Table 4, it can be seen that there are five criteria for testing the goodness of fit model that are in the good category. This means that the model in this study is very fit or a very good model.

Hypothesis test

Hypothesis testing is to test whether the research hypothesis can be accepted or rejected. Hypothesis testing in SEM analysis is carried out with a level of significance of 95% or α = 0.05. The hypothesis is accepted if the value of CR > 2 and has a p value < 0.05. For more details, it can be seen in Table 5 as follows.

Table 5 Research hypothesis test results

Based on Table 5, hypotheses 1 to 9 can be accepted. This study found that the better the UTAUT component accepted by consumers in digital payments usage, the higher the satisfaction felt by its users. Then, the user satisfaction that has been formed within the consumer is able to create inertia and CI in digital payment users.

Discussions

According to the statistical results presented above, this model supported the predictive validity of all the proposed factors. All factors match the main criteria of reliability and validity constructs. All loading factors are above 0.5, hence, it is suitable as a measurement parameter. This study model was able to explain, respectively, and approves the theoretical foundation. In some constructs of UTAUT2, the results of this study are in contrast to prior studies of Alalwan (2020) which discusses the measurement of the performance of the application of technology using UTAUT 2 which is able to have an impact on consumer satisfaction and CI on mobile food ordering apps. More specifically, UTAUT 2 in this study is divided into six constructs.

The first construct is performance expectancy. As seen in Table 2, PE was the most influential factor with 0.718 variance, in line with previous study (Alalwan 2020). This demonstrates the importance of the cognitive and functional benefits of digital payment from the customers’ perspective. The higher the performance expectancy provided by a digital payment platform, the more satisfied these users are. The second and third constructs are the effort expectancy and social influence. In contrast with Alalwan (2020) study in Jordan, this study shows that EE and SI influence user’s satisfaction positively. It might happen since the study was conducted in two different situations. The current study was carried out during the COVID-19 pandemic where people cannot leave the house as usual, practice social distancing, and are advised to use contactless tools, digital payment is the best away to protect oneself from contracting COVID-19 (Chotigo and Kadono 2021). Indonesian behavior is also different from Jordanian, where Indonesians are highly socialize and easily follow people they trust, especially in this case is baby boomer and X generation. In this COVID-19 pandemic, people also recommend and teach their family/parents to use digital payment to avoid contact and stay at home. In line with Chotigo and Kadono (2021) that Thailand customers’ perspectives regarding the use of digital apps during the COVID-19 pandemic seem to conform to the views with social support or the views of influencers, such as comments or reviews from social networks, advertising, and valuable references (family, friends, or colleagues) and Zhou (2011) the more digital payment platform users around them, the more likely that person will decide to use digital payment products which ultimately makes them feel satisfied because they used digital payments used by their social environment and feel up to date.

The fourth component is facilitating conditions. The existence of supporting facilities is capable of assisting the digital payment transaction process to be carried out properly—compatible smartphones for using digital payment applications, as well as Internet signal coverage. The better the facilitating conditions the user has, the more satisfied the user is with digital payment transactions because there are no obstacles that can prevent them from processing digital payment transactions. This result confirmed that of Alalwan (2020).

The fifth component is price value. The number of promotion (cashback/discounted/free shipping, etc.) and affordable transaction fee or even cheaper price when compared to cash payments make users feel satisfied with the performance of a digital payment platform, in line with Alalwan (2020). This logically justifies the important role of PV in shaping the customer’s satisfaction. The sixth component is habit. Habit is when users using digital payment transactions due to continuous behavior. The continuous use of digital payment transactions has become a habit and dependency. This dependence then makes users feel that digital payments are a part of their lives and in the end can create user satisfaction when they use it. This result is in line with Wang et al. (2013) and Alalwan (2020)

After measuring how users assess the performance of a digital payment platform and its impact on user satisfaction, we will then discuss how user satisfaction is able to create inertia and CI. The results show that the higher the level of user satisfaction with a technology, the higher the level of inertia felt by the user. This study is in line with previous studies (Limayem and Cheung 2008; Polites and Karahanna 2012; Wang et al. 2019). Digital payment customer satisfaction is able to create inertia which is a form of user attachment to this technology and will continue to use the same digital payment platform even though there are many other options. The success of user satisfaction creates this inertia that gives digital payment platforms an advantage because users are immune despite the large number of digital payment platform offers from competitors.

Furthermore, the results of this study also show that user satisfaction can affect the CI of digital payment platform users. The results of this study are supported by several previous studies that also found that the higher the level of user satisfaction with a technology, the higher the user’s intention to continue using the technology in a sustainable manner (Lin and Wang 2006; Bae 2018). The success of a digital payment platform in satisfying the users will be able to make users become loyal customers who will continue to be users of the service.

Inertia also has an impact on CI. In line with the results of previous studies (Kang and Lee 2010; Wang et al. 2019), this study found that inertia is able to create CI. The higher the inertia felt by users of a digital platform, the more likely it is that these users will use the digital platform on an ongoing basis. Inertia which is a form of attachment and persistence of a user to the use of a technology (Polites and Karahanna 2012) can be a basis for the formation of CI in digital payment.

Theoretical implications

This paper contributes to the advancement of theory regarding technology acceptance, as follows: First, this research is one of the initial attempts to explore the intention to use digital payment using UTAUT (Junadi 2015), how UTAUT affects satisfaction (Alalwan 2020), and the relationship between satisfaction, inertia, and CI (Wang et al. 2019). In other words, the three research models mentioned are integrated into a new simplified model to investigate the drivers of CI to digital payment. Venkatesh et al. (2012) empirically examined the validity of UTAUT2 to Hong Kong customer’s, a highly developed country, to explain their acceptance of mobile Internet services. Alalwan et al. (2017) examined Jordanian customers’ to explain their acceptance of mobile banking and MFOA using UTAUT2. Thus, this study is one of the forefront studies extending the applicability of UTAUT2 by examining new technology (digital payment) used by new users (baby boomers and X Generation) in new situation pandemic COVID-19 in developing country (Indonesia). Second, this study breaks the stereotype that the elderly generation tends to resist and difficult to accept technology, in line with (Magsamen-Conrad and Dillon 2020) that older adults are forced to technology in this pandemic. Hence, it might change their behavior to accept technology. Third, there are still limited studies on CI in digital payment. This study is to shed the light and adds the literature on CI in digital payment.

Managerial implications

The study adds new managerial insights and develops a vivid understanding of digital payment usage in baby boomers and X generation that will be beneficial for practitioners, digital payment providers, banks, financial institutions, or other mobile app providers and developers. The research results show that baby boomer and X generation will continue using digital payment after COVID-19 pandemic as they feel satisfied and inertia. It will become new segment and target market in digital payment or other digital services. It is important for bank, financial institution, or digital payment company to put more attention to the elderly as a lot of them start using digital payment from social influence who also taught them how to use it. Marketers must create marketing strategies that can reach the baby boomer and X generation, that are easy to understand, helpful, and also provide special treatment for them, such as teaching them how to use digital payments directly and recognize different users personality traits. In addition, companies are also advised to create a more user-friendly interface for the older generation, especially performance expectation is the most influential in this study and confirmed by prior studies (Alalwan 2020; Junadi 2015; Patil et al. 2017; Venkatesh et al. 2012). It is also undeniable that price value as one of the UTAUT2 constructs has also a big impact on customer satisfaction. Thus, the company is suggested to provide promotions regularly and keep costs affordable.

Limitations and future directions

The limitations of this study and directions for futures researchers are as follows: The first is there is a concern whether these results are generalizable, since most of the respondents in this study are older adults in the capital city of Indonesia who has higher educational background and acceptance of technology than people in the rural area. Second, this study was conducted at the time of the COVID-19 pandemic so that it allows compulsion of elderly consumers to use digital payments because of the situations, thus, it may be different in normal situations. Accordingly, future research can examine more about older adults behavior toward digital payment in normal situation after the pandemic ends. Third, future research should measure service quality to assess satisfaction and the impact of inertia to satisfaction. Researchers also can add construct Trust to extend UTAUT2 as was done by Alalwan et al. (2017), because trust is an important factor in the use of technology, especially those related to finance.

Conclusion

Our study contributes to theories of technology acceptance and adoption, particularly UTAUT model, and the continuance intention process as affected by user satisfaction and inertia across baby boomer and X generation. This is an addition to the fact that studies of older adults’ intention toward technology, particularly digital payment, are still lacking. This study realized the necessity of examining the main factors that could shape older adults’ intention to continue using digital payment after COVID-19 pandemic. By means of SEM methods, the authors find that UTAUT2 constructs facilitates users’ satisfaction, which in turn positively influences customers’ CI, with the highest construct of UTAUT2 that affects user’s satisfaction is performance expectancy with 0.728 variance and price value with 0.696 variance. Users’ satisfaction also positively influences inertia, and inertia positively influences CI. The findings show that most of them will continue using digital payment since they were accustomed to use it and feel satisfied. Overall, the authors provide a new perspective of the acceptance of technology by baby boomers and X generation.