Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

What Improves Customer Satisfaction in Mobile Banking Apps? An Application of Text Mining Analysis

What Improves Customer Satisfaction in Mobile Banking Apps? An Application of Text Mining Analysis Consumer-generated reviews reflect consumers' experiences and perceptions toward a product or service. In this context, we propose a text mining approach to identify factors that improve customer satisfaction in the mobile banking app service. To do so, we collect 96,140 mobile app reviews for four U.S. banks: Bank of America, Capital One, Chase, and Wells Fargo. Using the Latent Dirichlet Allocation (LDA) topic model, we first derive the critical quality dimensions such as ease of use, convenience, security, and customer support. Analysis of weekly panel data shows that positive responses to the security and convenience of mobile banking apps improve app ratings. However, increased comments about insecurity, negative customer support experiences, discomfort, and complexity lower user ratings. Overall, the empirical results support that security is the most influential factor in customer satisfaction with mobile financial services. Keywords: Mobile banking application, Financial services, Customer satisfaction, Text mining, Customer reviews customer experience and alleviate technical issues 1. Introduction using mobile banking applications (hereafter, apps). In this context, understanding which quality di- OVID-19 has accelerated digital trans- mensions facilitate or hinder user satisfaction with C formation in all industries around the world. mobile banking apps is an important research When it comes to using financial services, customers question. become reluctant to visit bank branches due to con- Mobile banking is one of the latest digital technol- cerns about the infection. Instead, a growing number ogies that combine mobcommunication technology of customers use Internet banking or mobile banking and financial services. Customers can perform as an alternative. Experts in the financial sector esti- various financial transactions through mobile apps mate that the pandemic dramatically accelerated that financial institutions provide. Customers can digital banking technology adoption (Mondres 2020). conveniently execute financial transactions such as Before the pandemic, retail banks had gradually balance inquiry, account transfer, and bill payment in adopted non-face-to-face channels such as ATMs real-time using smart devices (Shaikh and Karjaluoto and online banking to increase customer conve- 2015). Existing studies identify the factors affecting nience and efficiency. As the need for non-face-to- the acceptance intention of mobile banking based on face transactions proliferates, traditional banks face the technology acceptance model (hereafter, TAM). the challenges of increasing consumer satisfaction Customers using mobile banking apps generally and loyalty through mobile banking services (Shan- consider security, ease of use, and convenience are kar, Tiwari, and Gupta 2021). Conventional and on- essential (Sampaio, Ladeira, and Santini 2017). line-only banks should find ways to improve their This work was supported by the 2021 Research Fund of the University of Seoul. Received 23 December 2021; accepted 30 December 2021. Available online 3 February 2022. * Corresponding author. E-mail addresses: [email protected] (Y.K. Oh), [email protected] (J.-M. Kim). https://doi.org/10.53728/2765-6500.1581 2765-6500/© 2022 Korean Marketing Association (KMA). This is an open-access article under the CC-BY 4.0 license (https://creativecommons.org/licenses/by/4.0/). ASIA MARKETING JOURNAL 2022;23:28e37 29 Laukkanen (2007) studies the differences in customer literature related to the acceptance of mobile banking value perception between the Internet and mobile and conclude that perceived usefulness and compat- banking, suggesting that efficiency, convenience, and ibility with an individual's lifestyle are the main safety determine the main differences. Prior studies drivers of mobile banking acceptance. explore the quality dimensions that determine mo- bile banking service user satisfaction (Arcand et al. 2.2. Service quality dimensions of mobile banking 2017; Jun and Palacios 2016; Sampaio, Ladeira, and Santini 2017). Prior studies explore the quality dimensions that Consumers can leave reviews about their positive or explain customer satisfaction with mobile banking negative experiences with mobile banking apps. The apps. For example, Jun and Palacios (2016) identify vast amount of online reviews provide sources to un- mobile banking application quality and find that derstand reasons for customer reactions (Jeon et al. convenience, accuracy, diversity in features, ease of 2019; Leem and Eum 2021; Proctor 2021; Shankar, use, and continuous improvement are significant Tiwari, and Gupta 2021). Shankar, Tiwari, and Gupta influencers. Arcand et al. (2017) investigate the multi- (2021) conduct exploratory research by analyzing dimensional concept of mobile banking and find that mobile banking app reviews. They identify the critical utilitarian features such as security and practicality success factors for mobile banking: privacy/security, affect customer satisfaction mediated by customer navigation, customer support, convenience, and effi- trust. In addition, Sampaio, Ladeira, and Santini (2017) ciency. Although recent literature explores the major conduct survey research on the benefits offered by quality dimensions of mobile banking, no research mobile banking apps and find that the consequences examines how those quality dimensions affect user of satisfaction with mobile banking are trust, loyalty, ratings. This study derives the quality dimensions of and positive word-of-mouth. Thus, the previous mobile banking from a large scale of unstructured text studies on the service quality of mobile banking reviews. Furthermore, we examine which factors emphasize that a bank should prioritize the cus- significantly impact changes in user ratings. In tomers' perception of service value. particular, we analyze how the perception changes in the quality dimensions are related to the changes in 3. Conceptual framework and hypotheses customer satisfaction. Fig. 1 provides an overview of our conceptual The remainder of this paper is structured as follows. framework to explain how customer satisfaction The following section provides the theoretical back- changes in mobile banking apps. The selected ground and hypotheses on the quality dimensions in quality dimensions of the mobile banking service the mobile banking service based on the prior litera- are ease of use, convenience, security, and customer ture. We then explain the data and text mining anal- support. We expect that changes in the occurrence ysis process. Next, we describe the empirical model rate of service quality keywords will affect changes and report the hypothesis test results. Finally, we in user satisfaction. Next, we discuss the related suggest the implications of the main results and literature and develop hypotheses. discuss limitations and further research. 3.1. Ease of use 2. Theoretical background 2.1. Mobile banking adoption Many researchers reveal that perceived useful- ness and ease of use are the critical components for The technology acceptance model (TAM) based on accepting new IT-based services (Alalwan et al. Fishbein and Ajzen's(1977) theory assumes that users' 2016; Davis 1989; Gu, Lee, and Suh 2009; Kekre, beliefs and attitudes lead to behavioral intentions. Krishnan, and Srinivasan 1995; Sharma 2019). Gu, TAM has been widely used to understand the main Lee, and Suh (2009) propose that self-efficacy affects factors that affect mobile banking acceptance (Gu, Lee, mobile banking adoption through perceived ease of and Suh 2009; Sharma 2019). TAM theory argues that use. They suggest that banks develop a user- perceived usefulness and ease of use are fundamental friendly interface and quickly provide professional determinants of system adoption and use. Previous guidelines for mobile banking apps. Therefore, user studies on mobile banking acceptance identify influ- satisfaction will increase as more users easily navi- encing factors based on TAM. For example, Luo et al. gate and access various financial services in mobile (2010) show that perceived risks to mobile banking banking apps. Conversely, reduced perceived ease negatively affect consumers' intention to accept mo- of use due to the complexity of mobile app in- bile banking. Shaikh and Karjaluoto (2015) study the terfaces can reduce user satisfaction. 30 ASIA MARKETING JOURNAL 2022;23:28e37 Fig. 1. Conceptual framework. H1a. An increase (decrease) in the occurrence rate technical mistakes in the app cause delayed pay- of perceived ease of use in mobile banking app ment. Hence, technical problems such as crashes or reviews will be positively related to an increase connection errors related to using the mobile (decrease) in user satisfaction. banking app are likely to cause significant discom- forts to customers. H1b. An increase (decrease) in the occurrence rate H2a. An increase (decrease) in the occurrence rate of perceived complexity in mobile banking app of perceived convenience in mobile banking app reviews will be negatively related to an increase reviews will be positively related to an increase (decrease) in user satisfaction. (decrease) in user satisfaction. 3.2. Convenience H2b. An increase (decrease) in the occurrence rate of perceived inconvenience in mobile banking Service convenience refers to the characteristics of app reviews will be negatively related to an in- using the service with a minimum of time and effort crease (decrease) in user satisfaction. (Benoit, Klose, and Ettinger 2017). Convenience in mobile banking refers to the ability to conveniently 3.3. Security use the necessary financial services through the app anytime, anywhere (Shankar, Tiwari, and Gupta Security is one of the critical attributes valued by 2021). Consumers can transfer accounts using the customers conducting financial transactions mobile banking app without visiting a branch or through mobile banking apps (Sreejesh, Anusree, finding an ATM. In addition, mobile banking app and Amarnath 2016). Customers may be sensitive to users can receive notifications about bill payments security issues, such as whether their personal and to avoid overdue fees. Jebarajakirthy and Shankar financial transaction information is safe from hack- (2021) analyze the effect of multi-dimensions of ing. Perceived security concerns can be an essential convenience on mobile banking acceptance inten- reason users avoid financial transactions through tion. They reveal that access, transaction, search, online (Chang and Chen 2009). Recently, many and benefit conveniences are critical influencers on banks have introduced personal identification mobile banking service adoption. Therefore, con- technologies such as fingerprint recognition and venience is one of the essential factors for the pos- facial authentication into mobile apps to enhance itive evaluation of mobile banking apps. customer access convenience. Biometric authenti- In contrast, technical errors in a mobile banking cation is fast and straightforward, but it is also app make it challenging to use banking services. In vulnerable to hacker attacks. Therefore, banks particular, a high level of complaints may arise if should balance customer convenience with security. ASIA MARKETING JOURNAL 2022;23:28e37 31 H3a. An increase (decrease) in the occurrence rate causes the changes in consumer evaluation of mo- of perceived security in mobile banking app re- bile banking apps. Our data contains online review views will be positively related to an increase ratings and text data available at the Apple App (decrease) in user satisfaction. Store and Google Play Store. As of October 2021, the apps ranked by DAU (daily active users) are in the H3b. An increase (decrease) in the occurrence rate following order: Capital One, Chase, Bank of of perceived insecurity in mobile banking app America (BoA), and Wells Fargo. Capital One reviews will be negatively related to an increase operates as a specialized online bank without an (decrease) in user satisfaction. offline branch. The other three banks are traditional banks that operate physical offices and provide Internet and mobile banking services. 3.4. Customer support We select the reviews with 20 or more characters to identify the reasons for user satisfaction or Traditional banks have managed the quality of dissatisfaction. By doing so, the final sample in- face-to-face customer service to respond to various cludes 96,140 reviews. Table 1 presents descriptive needs related to customers' use of financial services. statistics for mobile app review characteristics for However, as the demand for non-face-to-face each bank. Capital One has the highest number of banking services increases, it is necessary to reviews (n ¼ 35,616) and the highest average rating respond to customer needs arising from new chan- (4.11 out of 5) during the sample period. BoA and nels. Specifically, banks should provide good Chase have lower mean ratings, higher standard customer support services that suggest solutions to deviations, longer review lengths, and a higher ratio customer inquiries related to mobile banking of negative opinions (1 or 2 points) than the other because financial services often offer very intensive two banks. In particular, the high variability of decisions such as money management (Jun and customer ratings suggests that user ratings are Palacios 2016; Shankar, Tiwari, and Gupta 2021). changing rather than static during the analysis When consumers solve problems through seam- period. These results support the notion that less communication, they can build trust in mobile achieving high levels of customer satisfaction is a apps and banks. Customer support related to mo- moving goal in the financial services industry bile banking apps includes answering questions (Krishnan et al. 1999). about financial services and troubleshooting tech- We illustrate the changes in the monthly ratio of nical errors such as mobile app crashes. However, positive reviews (5 out of 5) and negative reviews (1, consumers can discredit banks when they experi- 2 out of 5) in Fig. 2. In early 2019, Capital One's ence delayed contact with customer support or an mobile banking app had the highest favorable re- unfriendly attitude. Unpleasant experience with the views and the lowest negative reviews compared to customer support team may lead to dissatisfaction the other three banks. However, the percentage of and low ratings of the mobile banking apps. positive (negative) reviews on Capital One de- creases (increases) over time. Over the same period, H4a. An increase (decrease) in the occurrence rate three traditional banks (BoA, Chase, and Wells of pleasant customer support experience in mo- Fargo) appear to improve mobile banking service bile banking app reviews will be positively quality, closing the gap with the online-only bank related to an increase (decrease) in user satisfac- (Capital One). Next, we apply the text mining tion. technique to understand which factors improve/ deter customer satisfaction with mobile banking H4b. An increase (decrease) in the occurrence rate apps. of unpleasant customer support experience in mobile banking app reviews will be positively 4.2. Text mining analysis related to an increase (decrease) in user satisfaction. Text mining analysis of review data is similar to exploratory analysis of responses to open-ended 4. Empirical analysis questions (Allenby 2012). Our study uses the text mining approach in the following three steps: 4.1. Data natural language processing, topic model applica- tion, and explanatory variable construction. We We collect online reviews generated from then test the role of the identified components in February 2019 to October 2021 to understand what 32 ASIA MARKETING JOURNAL 2022;23:28e37 Table 1. Descriptive statistics of mobile banking app review characteristics. Company Obs. Rating Rating Review Length Positive_Ratio Negative_Ratio (mean) (stdev) (Rating ¼ 5) (Rating ¼ 1,2) BoA 16,184 3.17 1.80 189 0.44 0.42 CapitalOne 35,616 4.11 1.50 126 0.69 0.19 Chase 13,583 3.11 1.80 187 0.42 0.44 WellsFargo 30,757 3.97 1.58 126 0.66 0.23 Total 96,140 3.77 1.67 145 0.60 0.28 either promoting or degrading mobile banking the results of applying the topic model and related user ratings. literature. Step 1. Application of natural language processing Step 3. Construct explanatory variables for mobile banking app user rating In the preprocessing stage, we remove stop words and derive the words with high frequency: "bank,” Next, we measure the occurrence of mobile "account,” "payment,” "money,” "deposit,” "credit," banking service quality terms for each review. and "card." Except for the commonly appeared However, the terms might have opposite meanings banking terminologies, words that frequently if the selected keywords appear with negators (e.g., appear in positive reviews are "love,” "easy,” not, never, no, doesn't, isn't) in a phrase. To split the "convenient,” "user friendly,” "helpful,” "simple," text into phrases, we separate texts using punctua- and "fast." For the negative reviews, "update,” "fix,” tion marks (./,/:/;/ !/ ?) and conjunctions (e.g., and, "issue,” "crash,” "error message," and "white because, but, so) following the prior literature screen" appear frequently. (Büschken and Allenby 2020; Oh and Yi 2021). Then we exclude the phrases containing the selected Step 2. LDA topic modeling keywords and negators simultaneously. Table 2 also shows the occurrence rate and sample content of the This study applies the LDA (Latent Dirichlet constructs in the reviews for mobile banking apps. Allocation) topic model (Blei, Ng, and Jordan 2003), a type of unsupervised machine learning. The topic 4.3. Model estimation model is a technique that extracts various topics composed of a combination of specific words using We create a dummy variable indicating the the frequency of words appearing in the document. occurrence of selected keywords in a review after The topic model helps researchers extract key deriving quality dimensions for mobile banking quality dimensions reflected in a large scale of un- apps. A total of 44,141 reviews with at least one structured text data (Tirunillai and Tellis 2014). quality dimension are used for model estimation. Recent studies widely adopted the topic model to Then we build a weekly, bank-level panel data set extract latent topics in online reviews and elicit in- by taking the average of variables. Table 3 reports depth customer insights (Guo, Barnes, and Jia 2017; the descriptive statistics and correlation matrix for Leem and Eum 2021; Shankar, Tiwari, and Gupta the difference-in-difference of variables. 2021; Zhang 2019). The sample dataset has a cross-sectional and We apply the topic model to the reviews of posi- time-series format. Thus, we control for unobserved tive (Rating ¼ 5) and negative (Rating ¼ 1,2) sepa- bank-level heterogeneity by modeling the impact of rately. Then we identify the topics and top words changes in mobile banking quality dimensions on that contain attributes related to mobile banking in changes in user ratings (Luo and Homburg 2008; word combinations. Among the topics presented Luo, Homburg, and Wieseke 2010) as follows: through unsupervised learning, we recognize the words that appear in the topics related to ease of DRating ¼ a þ b Deasytouse þ b Dconvenience þ it 1 it 2 it use, convenience, security, and customer support. b Dsecurity þ b DCS PE þ b Dcomplexity it it it 3 4 5 Previous studies have also suggested that these variables are important for adopting mobile þ b Ddiscomfort þ b Dinsecurity þ b DCS NE þ e 6 it 7 it 8 it it banking services (Alalwan et al. 2016; Gu, Lee, and ð1Þ Suh 2009; Jebarajakirthy and Shankar 2021; Shankar, Tiwari, and Gupta 2021; Sharma 2019; Sreejesh, where DRating are changes in the mean rating of it Anusree, and Amarnath 2016). Table 2 summarizes bank i during week t. Deasytouse , Dconvenience , it it ASIA MARKETING JOURNAL 2022;23:28e37 33 Fig. 2. Changes in positive and negative review ratio over time. Dsecurity , and DCS PE are changes in the occur- significant effects except positive experience in it it rence rate of positive experience in ease of use, customer support (CS_PE ). The parameter estima- convenience, security, and customer support for tors in the fixed-effect model demonstrate that our bank i during week t, respectively. Dcomplexity constructs using the text mining approach have it Ddiscomfort , Dinsecurity and DCS NE are changes content validity to explain the aggregated customer it it it in the occurrence rate of negative experience in ease rating in mobile banking apps. of use, convenience, security, and customer support We estimate the first difference model using Eq. for bank i during week t. e is a residual term with (1) to investigate the factors that lead to changes in it variance s . customer response for mobile banking apps. We Additionally, we model the impact of changes in find that the increases in convenience and security are mobile banking quality dimensions on the ratio of associated with increases in Rating, PosR, and de- positive ratings (5 out of 5) and negative ratings (1,2 creases in NegR. In addition, increases in complexity, out of 5) as follows: discomfort, insecurity, and negative experience with customer support (CS_NE ) lead to declines in Rat- DPosR ¼ a þ b Deasytouse þ b Dconvenience þ it it it 1 2 ing, PosR, and increments in NegR. Our estimation results imply that changes in some positive quality b Dsecurity þ b DCS PE þ b Dcomplexity it it it 3 4 5 dimensions (easytouse and CS_PE ) may not affect þ b Ddiscomfort þ b Dinsecurity þ b DCS NE þ e it it it it 6 7 8 changes in customer reactions. However, increasing ð2Þ complaints regarding the key quality of mobile banking can hurt the overall ratings by increasing DNegR ¼ a þ b Deasytouse þ b Dconvenience þ it it it 1 2 negative feedback from customers. Our findings align with prior literature on the customer's asym- b Dsecurity þ b DCS PE þ b Dcomplexity it it it 3 4 5 metric reactions to service quality (Arbore and þ b Ddiscomfort þ b Dinsecurity þ b DCS NE þ e it it it it 6 7 8 Busacca 2009; Oh and Yi 2021). Further, we apply the first difference model to the ð3Þ bank subsample to examine what factors improved where, DPosR (DNegR Þ are changes in the ratio of it it or deterred the mobile banking app user experience positive (negative) ratings of bank i during week t. during the sample period. Estimation results (Table 5) imply that the improvements in Wells Fargo rat- 5. Results ing are mainly attributable to the increase in security, convenience, and easytouse. On the other hand, Cap- Table 4 shows the estimation results of the fixed ital One, which has the highest overall mean rating, effect model (level) and the first difference model experiences adverse reactions due to customer (change) for the weekly panel data. In the fixed-effect complaints regarding insecurity, CS_NE, discomfort, model, all constructs of mobile banking have signif- and complexity. The bank-level analysis shows that icant effects on the customer's overall rating (Rating) the quality dimensions of mobile banking apps have and the ratio of positive ratings (PosR) in the expected differential impacts on customer response from directions. In the negative ratio ratings (NegR) model, bank to bank. all explanatory variables show expected and 34 ASIA MARKETING JOURNAL 2022;23:28e37 Table 2. Text mining analysis of major quality dimensions in mobile banking. Construct Topic Terms Occurrence Content Sample Related Studies (%) Ease of use PE easy to use, simple, 66.15  The app is very easy to use. (Alalwan et al. straightforward,  This app gives me simplicity in 2016; Davis 1989; user-friendly, intuitive reviewing my account to check balances Gu, Lee, and Suh and pay off bills. 2009; Sharma 2019) NE hard, difficult, complicate 4.25  I am unsure why paying bills with the app got so difficult two weeks ago after the latest version or update. Most apps like this are so difficult and sensitive for security. Convenience PE convenient, track, notification, 26.02  Like how I can track and get payment (Benoit, Klose, and anytime, anywhere, fast, quick, alerts from text messages for approval Ettinger 2017; instant as to payment pending. Jebarajakirthy and Love the way I am instantly notified Shankar 2021; when a charge is posted to my account Shankar, Tiwari, by a merchant. and Gupta 2021) NE inconvenient, slow, crash, won't 18.17  A lot of crashes upon opening the app work, error, freeze, can't get, with several of these latest updates. white screen, please fix  I can't believe they still can't fix pay bills on the mobile app. Security PE security, secure 6.72  I feel totally secure using it to pay (Arcand et al. 2017; my payment. Chang and Chen I have touch id set up for the app, 2009; Shankar, which helps with security. Tiwari, and Gupta NE insecure, hacking, fraud, privacy, 1.06  I had fraudulent charges on my card. 2021; Sreejesh, concern  Constantly signing on today and it's Anusree, and making me leary if someone is Amarnath 2016) hacking my information. Customer PE {customer service, customer 3.61  Every time I talk to a customer service (Jun and Palacios Support support, staff, call} and {good, rep they are polite and very helpful. 2016; Shankar, great, excellent, best, helpful,  Customer service wait time is Tiwari, and Gupta friendly,polite} reasonable they are very friendly 2021) and helpful. NE {customer service, customer 0.81  Their customer service people are rude support, staff, call} and {poor, and don't know the first thing about worst, horrible, rude} true customer service. I've been a long time customer and have had the worst customer service over the years. Note: PE(NE) stands for positive(negative) user experience for the given construct. Table 3. Descriptive statistics and correlation matrix. DRating DPosR DNegR Deasytouse Dconvenience Dsecurity DCS_PE Dcomplexity Ddiscomfort Dinsecurity DCS_NE DRating 1.00 DPosR 0.91*** 1.00 DNegR 0.95*** 0.79*** 1.00 Deasytouse 0.30*** 0.30*** 0.30*** 1.00 Dconvenience 0.29*** 0.28*** 0.26*** 0.06 1.00 Dsecurity 0.31*** 0.28*** 0.30*** 0.06 0.05 1.00 DCS_PE 0.12*** 0.12** 0.10* 0.08 0.08 0.01 1.00 Dcomplexity 0.12*** 0.17*** 0.11** 0.20*** 0.14*** 0 0.06 1.00 Ddiscomfort 0.65*** 0.64*** 0.59*** 0.38*** 0.32*** 0.25*** 0.18*** 0.04 1.00 Dinsecurity 0.15*** 0.13** 0.15*** 0.09* 0.04 0.02 0.10* 0.01 0.12** 1.00 DCS_NE 0.12*** 0.11** 0.10* 0.10* 0.04 0.01 0.07 0.01 0.08* 0.01 1.00 Mean 0.004 0.001 0.001 0.0002 0.0004 0.0001 0.0004 0.0003 0.001 0.0002 0.0001 SD 0.44 0.12 0.12 0.1 0.09 0.05 0.04 0.05 0.12 0.03 0.03 Note: *p < 0.1; **p < 0.05; ***p < 0.01. ASIA MARKETING JOURNAL 2022;23:28e37 35 Table 4. Factors for improving and deterring mobile banking app user rating. Fixed Effect (Level) First Difference (Change) Rating PosR NegR DRating DPosR DNegR easytouse 0.50*** 0.12** 0.12** Deasytouse 0.10 0.01 0.06 (0.18) (0.05) (0.05) (0.15) (0.04) (0.04) convenience 0.91*** 0.21*** 0.21*** Dconvenience 0.45*** 0.09* 0.12** (0.21) (0.06) (0.06) (0.17) (0.05) (0.05) security 2.14*** 0.54*** 0.57*** Dsecurity 1.37*** 0.29*** 0.38*** (0.33) (0.10) (0.09) (0.27) (0.08) (0.08) CS_PE 0.77** 0.30*** 0.14 DCS_PE 0.53 0.12 0.10 (0.38) (0.11) (0.10) (0.34) (0.10) (0.10) complexity 2.56*** 0.85*** 0.58*** Dcomplexity 1.04*** 0.41*** 0.25*** (0.30) (0.09) (0.08) (0.26) (0.07) (0.08) discomfort 2.94*** 0.85*** 0.72*** Ddiscomfort 2.29*** 0.65*** 0.55*** (0.16) (0.05) (0.04) (0.15) (0.04) (0.04) insecurity 3.18*** 0.79*** 0.82*** Dinsecurity 3.33*** 0.87*** 0.84*** (0.54) (0.16) (0.15) (0.45) (0.13) (0.13) CS_NE 3.55*** 0.97*** 0.89*** DCS_NE 3.08*** 0.80*** 0.69*** (0.61) (0.18) (0.17) (0.50) (0.14) (0.15) Bank Fixed Effect Controlled Constant 0.004 0.001 0.001 (0.012) (0.004) (0.004) Obs. 569 569 569 565 565 565 R 0.80 0.78 0.77 0.55 0.53 0.47 Adj-R 0.80 0.78 0.76 0.54 0.52 0.46 Note: *p < 0.1; **p < 0.05; ***p < 0.01, Standard errors are in parentheses. Table 5. Factors for improving and deterring mobile banking app user rating by banks. Capital One Bank of America Chase Wells Fargo DRating DPosR DNegR DRating DRating DPosR DRating DPosR DNegR DRating DPosR DNegR Deasytouse 0.33 0.09 0.02 0.25 0.03 0.13 0.57 0.09 0.15 0.49* 0.03 0.21*** (0.24) (0.08) (0.07) (0.31) (0.09) (0.09) (0.35) (0.09) (0.11) (0.26) (0.08) (0.07) Dconvenience 0.53** 0.12 0.14* 1.04*** 0.30*** 0.22* 0.26 0.04 0.14 0.56** 0.06 0.21*** (0.24) (0.08) (0.07) (0.39) (0.11) (0.11) (0.37) (0.10) (0.11) (0.25) (0.07) (0.07) Dsecurity 0.55* 0.05 0.29*** 2.91*** 0.59*** 0.84*** 0.77 0.32** 0.05 0.79* 0.18 0.28** (0.33) (0.11) (0.10) (0.66) (0.19) (0.19) (0.60) (0.15) (0.18) (0.42) (0.12) (0.11) DCS_PE 0.37 0.06 0.14 1.48 0.32 0.29 0.22 0.17 0.05 0.55 0.15 0.16 (0.40) (0.14) (0.12) (0.95) (0.27) (0.28) (0.65) (0.17) (0.20) (0.63) (0.19) (0.17) Dcomplexity 2.07*** 0.63*** 0.49*** 0.52 0.24* 0.21 1.67*** 0.72*** 0.26 0.01 0.05 0.07 (0.39) (0.14) (0.12) (0.50) (0.14) (0.15) (0.56) (0.14) (0.17) (0.55) (0.16) (0.15) Ddiscomfort 2.80*** 0.79*** 0.72*** 2.03*** 0.62*** 0.44*** 2.20*** 0.52*** 0.58*** 2.36*** 0.70*** 0.56*** (0.22) (0.08) (0.07) (0.28) (0.08) (0.08) (0.36) (0.09) (0.11) (0.28) (0.08) (0.08) Dinsecurity 4.48*** 0.91*** 1.14*** 2.99*** 0.84*** 0.64** 3.32*** 0.90*** 0.93*** 3.63*** 1.16*** 0.79*** (0.59) (0.20) (0.18) (1.01) (0.29) (0.29) (0.88) (0.23) (0.27) (1.04) (0.31) (0.28) DCS_NE 4.23*** 0.96*** 1.12*** 2.68** 0.92*** 0.41 3.84*** 0.76*** 1.08*** 0.13 0.06 0.37 (0.69) (0.24) (0.21) (1.04) (0.30) (0.30) (0.99) (0.26) (0.30) (1.35) (0.40) (0.36) Constant 0.0003 0.0000 0.0002 0.004 0.0001 0.001 0.01 0.002 0.003 0.01 0.002 0.001 (0.01) (0.005) (0.004) (0.03) (0.01) (0.01) (0.03) (0.01) (0.01) (0.02) (0.01) (0.005) Obs. 136 136 136 143 143 143 143 143 143 143 143 143 R 0.78 0.68 0.73 0.63 0.61 0.53 0.39 0.43 0.29 0.66 0.60 0.67 Adj-R 0.77 0.66 0.71 0.60 0.59 0.51 0.35 0.40 0.25 0.64 0.58 0.65 Note: *p < 0.1; **p < 0.05; ***p < 0.01, Standard errors are in parentheses. do so, we collect 96,140 mobile app reviews for four 6. Conclusions U.S. banks: Bank of America, Capital One, Chase, 6.1. Summary and Wells Fargo. We first extract quality dimensions using the LDA topic model and interpret the factors This study examines the factors to improve as ease of use, convenience, security, and customer customer satisfaction in mobile banking services. To 36 ASIA MARKETING JOURNAL 2022;23:28e37 support based on the prior literature. We then (Shankar and Jebarajakirthy 2019). Thus, increasing conduct panel data analysis to investigate how user satisfaction with mobile banking apps is changes in each factor affect user ratings. Our esti- essential to retain loyal consumers and prevent mation results show that user ratings get improved customer churn. In particular, reducing complaints as positive responses to security and convenience due to poor customer support and technical errors is increase. However, user rating declines as reactions critical. Therefore, banks should utilize marketing to insecurity, negative experience with customer intelligence systems to monitor changing customer support, discomfort, and complexity accumulate. reactions reflected in user-generated reviews. Overall, we find that security is the most influential factor that affects user ratings. Moreover, analysis 6.4. Limitations and future research results for each bank suggest that the effect of each factor on customer satisfaction may differ from bank It is challenging to determine whether a phrase to bank. with negators means positive or negative until reading the actual review. Hence, this study con- 6.2. Theoretical implications structs and analyzes samples after removing phra- ses containing negators. However, the number of This study contributes to the existing literature in sample reviews will increase if advanced text min- the following two aspects. First, this study applies ing techniques correctly classify sentences with ne- text mining analysis to online reviews to explain gators. In such a case, the accuracy of detecting users' satisfaction with mobile banking services. As positive/negative perceptions of mobile banking the importance of mobile platforms increases, many service quality is also likely to increase. recent studies have investigated mobile app reviews This study does not examine banking-related to understand the motives of consumer behavior functions such as deposit, money transfer, bill pay- (Liu et al. 2019; Verkijika and Neneh 2021). How- ment, and credit score management. In contrast, this ever, only a few recent studies explore major quality study focuses on the four quality dimensions of dimensions of mobile banking app with customer- mobile banking service: ease of use, convenience, generated reviews (Leem and Eum 2021; Shankar, security, and customer support. With the merit of Tiwari, and Gupta 2021). Our study extends the rich consumer-generated text data, a future study prior research by identifying major quality di- can analyze how users' evaluation of each functional mensions using natural language processing and element affects mobile banking app satisfaction. machine learning methods. Research on customer response using online re- Second, this study is the first paper to study how views and text mining can provide in-depth insights the factors extracted through text analysis on online into consumer behavior towards financial services. reviews affect users' satisfaction. Existing studies related to mobile banking acceptance intention References primarily conduct a survey approach and identify influencing factors at a certain point in time. Unlike Alalwan, Ali Abdallah, Yogesh K. Dwivedi, Nripendra P.P. Rana, prior studies, we establish weekly panel data using and Michael D. Williams (2016), “Consumer Adoption of Mobile Banking in Jordan: Examining the Role of Usefulness, large-scale reviews to investigate the dynamic effect Ease of Use, Perceived Risk and Self-Efficacy,” Journal of En- of service quality factors on customer satisfaction. terprise Information Management, 29 (1), 118e39. Allenby, Greg (2012), “Keynote Speech at the Fall 2012 KMA Conference: Big Data 2.0,” Asia Marketing Journal, 14 (3), 1e5. 6.3. Managerial implications Arbore, Alessandro and Bruno Busacca (2009), “Customer Satis- faction and Dissatisfaction in Retail Banking: Exploring the The results of this study have the following im- Asymmetric Impact of Attribute Performances,” Journal of Retailing and Consumer Services, 16 (4), 271e80. plications. Security has the most significant influ- Arcand, Manon, Sandrine PromTep, Isabelle Brun, and ence on user satisfaction. In online reviews, words Lova Rajaobelina (2017), “Mobile Banking Service Quality and related to ease of use or convenience are left 4 to 10 Customer Relationships,” International Journal of Bank Market- ing, 35 (7), 1068e89. times more than words related to security. However, Benoit, Sabine, Sonja Klose, and Andreas Ettinger. (2017), user satisfaction with security has a greater impact “Linking Service Convenience to Satisfaction: Dimensions and on user ratings. These results suggest that banks Key Moderators,” Journal of Services Marketing, 31 (6), 527e38. Blei, David M., Andrew Y. Ng, and Michael I. Jordan (2003), need to pay special attention to security to increase “Latent Dirichlet Allocation,” Journal of Machine Learning customer satisfaction when designing mobile Research, 3, 993e1022. banking apps. Büschken, Joachim and Greg M. Allenby (2020), “Improving Text Analysis Using Sentence Conjunctions and Punctuation,” Currently, banks face increasingly fierce compe- Marketing Science, 39 (4), 727e42. tition to provide a better mobile services experience ASIA MARKETING JOURNAL 2022;23:28e37 37 Chang, Hsin Hsin and Su Wen Chen (2009), “Consumer Acceptance of Emerging Technologies: An Empirical Study of Perception of Interface Quality, Security, and Loyalty in Mobile Banking Services,” Decision Support Systems, 49 (2), Electronic Commerce,” Information & Management, 46 (7), 222e34. 411e7. Luo, Xueming, Christian Homburg, and Wieseke Jan (2010), Davis, Fred D. (1989), “Perceived Usefulness, Perceived Ease of “Customer Satisfaction, Analyst Stock Recommendations, and Use, and User Acceptance of Information Technology,” MIS Firm Value,” Journal of Marketing Research, 47 (6), 1041e58. Quarterly, 13 (3), 319e40. Mondres. (2020), “ABA Data Bank: Mobile Banking Adoption Fishbein, Martin and Icek Ajzen (1977), “Belief, Attitude, Inten- Accelerates,” ABA Banking Journal. Available at:https:// tion, and Behavior: An Introduction to Theory and Research,” bankingjournal.aba.com/2020/10/aba-data-bank-mobile- Philosophy and Rhetoric, 10 (2), 130e2. banking-adoption-accelerates/ . (accessed December 2, 2021). Gu, Ja-Chul, Sang-Chul Lee, and Yung-Ho Suh (2009), “De- Oh, Yun Kyung and Jisu Yi (2021), “Asymmetric Effect of Feature terminants of Behavioral Intention to Mobile Banking,” Expert Level Sentiment on Product Rating: An Application of Bigram Systems with Applications, 36 (9), 11605e16. Natural Language Processing (NLP) Analysis,” Internet Guo, Yue, Stuart J. Barnes, and Qiong Jia (2017), “Mining Research, https://doi.org/10.1108/INTR-11-2020-0649 Meaning from Online Ratings and Reviews: Tourist Satisfac- Proctor, Jasmine (2021), “Labour of Love: Fan Labour, BTS, and tion Analysis Using Latent Dirichlet Allocation,” Tourism South Korean Soft Power,” Asia Marketing Journal, 22 (4), Management, 59, 467e83. 79e101. Jebarajakirthy, Charles and Amit Shankar (2021), “Impact of Sampaio, Claudio Hoffmann, Wagner Junior Ladeira, and Online Convenience on Mobile Banking Adoption Intention: Fernando De Oliveira Santini (2017), “Apps for Mobile A Moderated Mediation Approach,” Journal of Retailing and Banking and Customer Satisfaction: A Cross-Cultural Study,” Consumer Services, 58, 102323. International Journal of Bank Marketing, 35 (7), 1133e53. Jeon, Jaihyun, Taewook Lim, Byung-Do Kim, and Junhee Seok Shaikh, Aijaz A. and Heikki Karjaluoto (2015), “Mobile Banking (2019), “Effect of Online Word of Mouth on Product Sales: Adoption: A Literature Review,” Telematics and Informatics,32 Focusing on Communication-Channel Characteristics,” Asia (1), 129e42. Marketing Journal, 21 (2), 73e98. Shankar, Amit and Charles Jebarajakirthy (2019), “The Influence Jun, Minjoon and Sergio Palacios (2016), “Examining the Key of E-Banking Service Quality on Customer Loyalty: A Dimensions of Mobile Banking Service Quality: An Explor- Moderated Mediation Approach,” International Journal of Bank atory Study,” International Journal of Bank Marketing, 34 (3), Marketing, 37 (5), 1119e42. 307e26. Shankar, Amit, Aviral Kumar Tiwari, and Manish Gupta (2021), Kekre, Sunder, Mayuram S. Krishnan, and Kannan Srinivasan “Sustainable Mobile Banking Application: A Text Mining (1995), “Drivers of Customer Satisfaction for Software Prod- Approach to Explore Critical Success Factors,” Journal of En- ucts: Implications for Design and Service Support,” Manage- terprise Information Management, https://doi.org/10.1108/JEIM- ment Science, 41 (9), 1456e70. 10-2020-0426 Krishnan, M.S., Venkatram Ramaswamy, Mary C. Meyer, and Sharma, Sujeet Kumar (2019), “Integrating Cognitive Antecedents Damien Paul (1999), “Customer Satisfaction for Financial into TAM to Explain Mobile Banking Behavioral Intention: A Services: The Role of Products, Services, and Information SEM-Neural Network Modeling,” Information Systems Frontiers, Technology,” Management Science, 45 (9), 1194e209. 21 (4), 815e27. Laukkanen, Tommi (2007), “Internet vs Mobile Banking: Sreejesh, S., M.R. Anusree, and Amarnath Mitra (2016), “Effect of Comparing Customer Value Perceptions,” Business Process Information Content and Form on Customers’ Attitude and Management Journal, 13 (6), 788e97. Transaction Intention in Mobile Banking: Moderating Role of Leem, Byung-Hak and Seong-Won Eum (2021), “Using Text Perceived Privacy Concern,” International Journal of Bank Mining to Measure Mobile Banking Service Quality,” Indus- Marketing, 34 (7), 1092e113. trial Management & Data Systems, 121 (5), 993e1007. Tirunillai, Seshadri and Gerard J. Tellis (2014), “Mining Market- Liu, Hongfei, Chanaka Jayawardhena, Sally Dibb, and ing Meaning from Online Chatter: Strategic Brand Analysis of Chatura Ranaweera (2019), “Examining the Trade-Off be- Big Data Using Latent Dirichlet Allocation,” Journal of Mar- tween Compensation and Promptness in eWOM-Triggered keting Research, 51 (4), 463e79. Service Recovery: A Restorative Justice Perspective,” Tourism Verkijika, Silas Formunyuy and Brownhilder Ngek Neneh (2021), Management, 75, 381e92. “Standing Up For or Against: A Text-Mining Study on the Luo, Xueming and Christian Homburg (2008), “Satisfaction, Recommendation of Mobile Payment Apps,” Journal of Complaint, and the Stock Value Gap,” Journal of Marketing,72 Retailing and Consumer Services, 63, 102743. (4), 29e43. Zhang, Jurui (2019), “What’s Yours is Mine: Exploring Customer Luo, Xin, Li Han, Jie Zhang, and J.P. Shim (2010), “Examining Voice on Airbnb Using Text-Mining Approaches,” Journal of Multi-Dimensional Trust and Multi-Faceted Risk in Initial Consumer Marketing, 36 (5), 655e65. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Asia Marketing Journal Unpaywall

What Improves Customer Satisfaction in Mobile Banking Apps? An Application of Text Mining Analysis

Oh, Yun Kyung; Kim, Jung-Min
Asia Marketing Journal – Feb 4, 2022
Download PDF
11 pages

Loading next page...
 
/lp/unpaywall/what-improves-customer-satisfaction-in-mobile-banking-apps-an-Ts6Pj1GB27

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
2765-6500
DOI
10.53728/2765-6500.1581
Publisher site
See Article on Publisher Site

Abstract

Consumer-generated reviews reflect consumers' experiences and perceptions toward a product or service. In this context, we propose a text mining approach to identify factors that improve customer satisfaction in the mobile banking app service. To do so, we collect 96,140 mobile app reviews for four U.S. banks: Bank of America, Capital One, Chase, and Wells Fargo. Using the Latent Dirichlet Allocation (LDA) topic model, we first derive the critical quality dimensions such as ease of use, convenience, security, and customer support. Analysis of weekly panel data shows that positive responses to the security and convenience of mobile banking apps improve app ratings. However, increased comments about insecurity, negative customer support experiences, discomfort, and complexity lower user ratings. Overall, the empirical results support that security is the most influential factor in customer satisfaction with mobile financial services. Keywords: Mobile banking application, Financial services, Customer satisfaction, Text mining, Customer reviews customer experience and alleviate technical issues 1. Introduction using mobile banking applications (hereafter, apps). In this context, understanding which quality di- OVID-19 has accelerated digital trans- mensions facilitate or hinder user satisfaction with C formation in all industries around the world. mobile banking apps is an important research When it comes to using financial services, customers question. become reluctant to visit bank branches due to con- Mobile banking is one of the latest digital technol- cerns about the infection. Instead, a growing number ogies that combine mobcommunication technology of customers use Internet banking or mobile banking and financial services. Customers can perform as an alternative. Experts in the financial sector esti- various financial transactions through mobile apps mate that the pandemic dramatically accelerated that financial institutions provide. Customers can digital banking technology adoption (Mondres 2020). conveniently execute financial transactions such as Before the pandemic, retail banks had gradually balance inquiry, account transfer, and bill payment in adopted non-face-to-face channels such as ATMs real-time using smart devices (Shaikh and Karjaluoto and online banking to increase customer conve- 2015). Existing studies identify the factors affecting nience and efficiency. As the need for non-face-to- the acceptance intention of mobile banking based on face transactions proliferates, traditional banks face the technology acceptance model (hereafter, TAM). the challenges of increasing consumer satisfaction Customers using mobile banking apps generally and loyalty through mobile banking services (Shan- consider security, ease of use, and convenience are kar, Tiwari, and Gupta 2021). Conventional and on- essential (Sampaio, Ladeira, and Santini 2017). line-only banks should find ways to improve their This work was supported by the 2021 Research Fund of the University of Seoul. Received 23 December 2021; accepted 30 December 2021. Available online 3 February 2022. * Corresponding author. E-mail addresses: [email protected] (Y.K. Oh), [email protected] (J.-M. Kim). https://doi.org/10.53728/2765-6500.1581 2765-6500/© 2022 Korean Marketing Association (KMA). This is an open-access article under the CC-BY 4.0 license (https://creativecommons.org/licenses/by/4.0/). ASIA MARKETING JOURNAL 2022;23:28e37 29 Laukkanen (2007) studies the differences in customer literature related to the acceptance of mobile banking value perception between the Internet and mobile and conclude that perceived usefulness and compat- banking, suggesting that efficiency, convenience, and ibility with an individual's lifestyle are the main safety determine the main differences. Prior studies drivers of mobile banking acceptance. explore the quality dimensions that determine mo- bile banking service user satisfaction (Arcand et al. 2.2. Service quality dimensions of mobile banking 2017; Jun and Palacios 2016; Sampaio, Ladeira, and Santini 2017). Prior studies explore the quality dimensions that Consumers can leave reviews about their positive or explain customer satisfaction with mobile banking negative experiences with mobile banking apps. The apps. For example, Jun and Palacios (2016) identify vast amount of online reviews provide sources to un- mobile banking application quality and find that derstand reasons for customer reactions (Jeon et al. convenience, accuracy, diversity in features, ease of 2019; Leem and Eum 2021; Proctor 2021; Shankar, use, and continuous improvement are significant Tiwari, and Gupta 2021). Shankar, Tiwari, and Gupta influencers. Arcand et al. (2017) investigate the multi- (2021) conduct exploratory research by analyzing dimensional concept of mobile banking and find that mobile banking app reviews. They identify the critical utilitarian features such as security and practicality success factors for mobile banking: privacy/security, affect customer satisfaction mediated by customer navigation, customer support, convenience, and effi- trust. In addition, Sampaio, Ladeira, and Santini (2017) ciency. Although recent literature explores the major conduct survey research on the benefits offered by quality dimensions of mobile banking, no research mobile banking apps and find that the consequences examines how those quality dimensions affect user of satisfaction with mobile banking are trust, loyalty, ratings. This study derives the quality dimensions of and positive word-of-mouth. Thus, the previous mobile banking from a large scale of unstructured text studies on the service quality of mobile banking reviews. Furthermore, we examine which factors emphasize that a bank should prioritize the cus- significantly impact changes in user ratings. In tomers' perception of service value. particular, we analyze how the perception changes in the quality dimensions are related to the changes in 3. Conceptual framework and hypotheses customer satisfaction. Fig. 1 provides an overview of our conceptual The remainder of this paper is structured as follows. framework to explain how customer satisfaction The following section provides the theoretical back- changes in mobile banking apps. The selected ground and hypotheses on the quality dimensions in quality dimensions of the mobile banking service the mobile banking service based on the prior litera- are ease of use, convenience, security, and customer ture. We then explain the data and text mining anal- support. We expect that changes in the occurrence ysis process. Next, we describe the empirical model rate of service quality keywords will affect changes and report the hypothesis test results. Finally, we in user satisfaction. Next, we discuss the related suggest the implications of the main results and literature and develop hypotheses. discuss limitations and further research. 3.1. Ease of use 2. Theoretical background 2.1. Mobile banking adoption Many researchers reveal that perceived useful- ness and ease of use are the critical components for The technology acceptance model (TAM) based on accepting new IT-based services (Alalwan et al. Fishbein and Ajzen's(1977) theory assumes that users' 2016; Davis 1989; Gu, Lee, and Suh 2009; Kekre, beliefs and attitudes lead to behavioral intentions. Krishnan, and Srinivasan 1995; Sharma 2019). Gu, TAM has been widely used to understand the main Lee, and Suh (2009) propose that self-efficacy affects factors that affect mobile banking acceptance (Gu, Lee, mobile banking adoption through perceived ease of and Suh 2009; Sharma 2019). TAM theory argues that use. They suggest that banks develop a user- perceived usefulness and ease of use are fundamental friendly interface and quickly provide professional determinants of system adoption and use. Previous guidelines for mobile banking apps. Therefore, user studies on mobile banking acceptance identify influ- satisfaction will increase as more users easily navi- encing factors based on TAM. For example, Luo et al. gate and access various financial services in mobile (2010) show that perceived risks to mobile banking banking apps. Conversely, reduced perceived ease negatively affect consumers' intention to accept mo- of use due to the complexity of mobile app in- bile banking. Shaikh and Karjaluoto (2015) study the terfaces can reduce user satisfaction. 30 ASIA MARKETING JOURNAL 2022;23:28e37 Fig. 1. Conceptual framework. H1a. An increase (decrease) in the occurrence rate technical mistakes in the app cause delayed pay- of perceived ease of use in mobile banking app ment. Hence, technical problems such as crashes or reviews will be positively related to an increase connection errors related to using the mobile (decrease) in user satisfaction. banking app are likely to cause significant discom- forts to customers. H1b. An increase (decrease) in the occurrence rate H2a. An increase (decrease) in the occurrence rate of perceived complexity in mobile banking app of perceived convenience in mobile banking app reviews will be negatively related to an increase reviews will be positively related to an increase (decrease) in user satisfaction. (decrease) in user satisfaction. 3.2. Convenience H2b. An increase (decrease) in the occurrence rate of perceived inconvenience in mobile banking Service convenience refers to the characteristics of app reviews will be negatively related to an in- using the service with a minimum of time and effort crease (decrease) in user satisfaction. (Benoit, Klose, and Ettinger 2017). Convenience in mobile banking refers to the ability to conveniently 3.3. Security use the necessary financial services through the app anytime, anywhere (Shankar, Tiwari, and Gupta Security is one of the critical attributes valued by 2021). Consumers can transfer accounts using the customers conducting financial transactions mobile banking app without visiting a branch or through mobile banking apps (Sreejesh, Anusree, finding an ATM. In addition, mobile banking app and Amarnath 2016). Customers may be sensitive to users can receive notifications about bill payments security issues, such as whether their personal and to avoid overdue fees. Jebarajakirthy and Shankar financial transaction information is safe from hack- (2021) analyze the effect of multi-dimensions of ing. Perceived security concerns can be an essential convenience on mobile banking acceptance inten- reason users avoid financial transactions through tion. They reveal that access, transaction, search, online (Chang and Chen 2009). Recently, many and benefit conveniences are critical influencers on banks have introduced personal identification mobile banking service adoption. Therefore, con- technologies such as fingerprint recognition and venience is one of the essential factors for the pos- facial authentication into mobile apps to enhance itive evaluation of mobile banking apps. customer access convenience. Biometric authenti- In contrast, technical errors in a mobile banking cation is fast and straightforward, but it is also app make it challenging to use banking services. In vulnerable to hacker attacks. Therefore, banks particular, a high level of complaints may arise if should balance customer convenience with security. ASIA MARKETING JOURNAL 2022;23:28e37 31 H3a. An increase (decrease) in the occurrence rate causes the changes in consumer evaluation of mo- of perceived security in mobile banking app re- bile banking apps. Our data contains online review views will be positively related to an increase ratings and text data available at the Apple App (decrease) in user satisfaction. Store and Google Play Store. As of October 2021, the apps ranked by DAU (daily active users) are in the H3b. An increase (decrease) in the occurrence rate following order: Capital One, Chase, Bank of of perceived insecurity in mobile banking app America (BoA), and Wells Fargo. Capital One reviews will be negatively related to an increase operates as a specialized online bank without an (decrease) in user satisfaction. offline branch. The other three banks are traditional banks that operate physical offices and provide Internet and mobile banking services. 3.4. Customer support We select the reviews with 20 or more characters to identify the reasons for user satisfaction or Traditional banks have managed the quality of dissatisfaction. By doing so, the final sample in- face-to-face customer service to respond to various cludes 96,140 reviews. Table 1 presents descriptive needs related to customers' use of financial services. statistics for mobile app review characteristics for However, as the demand for non-face-to-face each bank. Capital One has the highest number of banking services increases, it is necessary to reviews (n ¼ 35,616) and the highest average rating respond to customer needs arising from new chan- (4.11 out of 5) during the sample period. BoA and nels. Specifically, banks should provide good Chase have lower mean ratings, higher standard customer support services that suggest solutions to deviations, longer review lengths, and a higher ratio customer inquiries related to mobile banking of negative opinions (1 or 2 points) than the other because financial services often offer very intensive two banks. In particular, the high variability of decisions such as money management (Jun and customer ratings suggests that user ratings are Palacios 2016; Shankar, Tiwari, and Gupta 2021). changing rather than static during the analysis When consumers solve problems through seam- period. These results support the notion that less communication, they can build trust in mobile achieving high levels of customer satisfaction is a apps and banks. Customer support related to mo- moving goal in the financial services industry bile banking apps includes answering questions (Krishnan et al. 1999). about financial services and troubleshooting tech- We illustrate the changes in the monthly ratio of nical errors such as mobile app crashes. However, positive reviews (5 out of 5) and negative reviews (1, consumers can discredit banks when they experi- 2 out of 5) in Fig. 2. In early 2019, Capital One's ence delayed contact with customer support or an mobile banking app had the highest favorable re- unfriendly attitude. Unpleasant experience with the views and the lowest negative reviews compared to customer support team may lead to dissatisfaction the other three banks. However, the percentage of and low ratings of the mobile banking apps. positive (negative) reviews on Capital One de- creases (increases) over time. Over the same period, H4a. An increase (decrease) in the occurrence rate three traditional banks (BoA, Chase, and Wells of pleasant customer support experience in mo- Fargo) appear to improve mobile banking service bile banking app reviews will be positively quality, closing the gap with the online-only bank related to an increase (decrease) in user satisfac- (Capital One). Next, we apply the text mining tion. technique to understand which factors improve/ deter customer satisfaction with mobile banking H4b. An increase (decrease) in the occurrence rate apps. of unpleasant customer support experience in mobile banking app reviews will be positively 4.2. Text mining analysis related to an increase (decrease) in user satisfaction. Text mining analysis of review data is similar to exploratory analysis of responses to open-ended 4. Empirical analysis questions (Allenby 2012). Our study uses the text mining approach in the following three steps: 4.1. Data natural language processing, topic model applica- tion, and explanatory variable construction. We We collect online reviews generated from then test the role of the identified components in February 2019 to October 2021 to understand what 32 ASIA MARKETING JOURNAL 2022;23:28e37 Table 1. Descriptive statistics of mobile banking app review characteristics. Company Obs. Rating Rating Review Length Positive_Ratio Negative_Ratio (mean) (stdev) (Rating ¼ 5) (Rating ¼ 1,2) BoA 16,184 3.17 1.80 189 0.44 0.42 CapitalOne 35,616 4.11 1.50 126 0.69 0.19 Chase 13,583 3.11 1.80 187 0.42 0.44 WellsFargo 30,757 3.97 1.58 126 0.66 0.23 Total 96,140 3.77 1.67 145 0.60 0.28 either promoting or degrading mobile banking the results of applying the topic model and related user ratings. literature. Step 1. Application of natural language processing Step 3. Construct explanatory variables for mobile banking app user rating In the preprocessing stage, we remove stop words and derive the words with high frequency: "bank,” Next, we measure the occurrence of mobile "account,” "payment,” "money,” "deposit,” "credit," banking service quality terms for each review. and "card." Except for the commonly appeared However, the terms might have opposite meanings banking terminologies, words that frequently if the selected keywords appear with negators (e.g., appear in positive reviews are "love,” "easy,” not, never, no, doesn't, isn't) in a phrase. To split the "convenient,” "user friendly,” "helpful,” "simple," text into phrases, we separate texts using punctua- and "fast." For the negative reviews, "update,” "fix,” tion marks (./,/:/;/ !/ ?) and conjunctions (e.g., and, "issue,” "crash,” "error message," and "white because, but, so) following the prior literature screen" appear frequently. (Büschken and Allenby 2020; Oh and Yi 2021). Then we exclude the phrases containing the selected Step 2. LDA topic modeling keywords and negators simultaneously. Table 2 also shows the occurrence rate and sample content of the This study applies the LDA (Latent Dirichlet constructs in the reviews for mobile banking apps. Allocation) topic model (Blei, Ng, and Jordan 2003), a type of unsupervised machine learning. The topic 4.3. Model estimation model is a technique that extracts various topics composed of a combination of specific words using We create a dummy variable indicating the the frequency of words appearing in the document. occurrence of selected keywords in a review after The topic model helps researchers extract key deriving quality dimensions for mobile banking quality dimensions reflected in a large scale of un- apps. A total of 44,141 reviews with at least one structured text data (Tirunillai and Tellis 2014). quality dimension are used for model estimation. Recent studies widely adopted the topic model to Then we build a weekly, bank-level panel data set extract latent topics in online reviews and elicit in- by taking the average of variables. Table 3 reports depth customer insights (Guo, Barnes, and Jia 2017; the descriptive statistics and correlation matrix for Leem and Eum 2021; Shankar, Tiwari, and Gupta the difference-in-difference of variables. 2021; Zhang 2019). The sample dataset has a cross-sectional and We apply the topic model to the reviews of posi- time-series format. Thus, we control for unobserved tive (Rating ¼ 5) and negative (Rating ¼ 1,2) sepa- bank-level heterogeneity by modeling the impact of rately. Then we identify the topics and top words changes in mobile banking quality dimensions on that contain attributes related to mobile banking in changes in user ratings (Luo and Homburg 2008; word combinations. Among the topics presented Luo, Homburg, and Wieseke 2010) as follows: through unsupervised learning, we recognize the words that appear in the topics related to ease of DRating ¼ a þ b Deasytouse þ b Dconvenience þ it 1 it 2 it use, convenience, security, and customer support. b Dsecurity þ b DCS PE þ b Dcomplexity it it it 3 4 5 Previous studies have also suggested that these variables are important for adopting mobile þ b Ddiscomfort þ b Dinsecurity þ b DCS NE þ e 6 it 7 it 8 it it banking services (Alalwan et al. 2016; Gu, Lee, and ð1Þ Suh 2009; Jebarajakirthy and Shankar 2021; Shankar, Tiwari, and Gupta 2021; Sharma 2019; Sreejesh, where DRating are changes in the mean rating of it Anusree, and Amarnath 2016). Table 2 summarizes bank i during week t. Deasytouse , Dconvenience , it it ASIA MARKETING JOURNAL 2022;23:28e37 33 Fig. 2. Changes in positive and negative review ratio over time. Dsecurity , and DCS PE are changes in the occur- significant effects except positive experience in it it rence rate of positive experience in ease of use, customer support (CS_PE ). The parameter estima- convenience, security, and customer support for tors in the fixed-effect model demonstrate that our bank i during week t, respectively. Dcomplexity constructs using the text mining approach have it Ddiscomfort , Dinsecurity and DCS NE are changes content validity to explain the aggregated customer it it it in the occurrence rate of negative experience in ease rating in mobile banking apps. of use, convenience, security, and customer support We estimate the first difference model using Eq. for bank i during week t. e is a residual term with (1) to investigate the factors that lead to changes in it variance s . customer response for mobile banking apps. We Additionally, we model the impact of changes in find that the increases in convenience and security are mobile banking quality dimensions on the ratio of associated with increases in Rating, PosR, and de- positive ratings (5 out of 5) and negative ratings (1,2 creases in NegR. In addition, increases in complexity, out of 5) as follows: discomfort, insecurity, and negative experience with customer support (CS_NE ) lead to declines in Rat- DPosR ¼ a þ b Deasytouse þ b Dconvenience þ it it it 1 2 ing, PosR, and increments in NegR. Our estimation results imply that changes in some positive quality b Dsecurity þ b DCS PE þ b Dcomplexity it it it 3 4 5 dimensions (easytouse and CS_PE ) may not affect þ b Ddiscomfort þ b Dinsecurity þ b DCS NE þ e it it it it 6 7 8 changes in customer reactions. However, increasing ð2Þ complaints regarding the key quality of mobile banking can hurt the overall ratings by increasing DNegR ¼ a þ b Deasytouse þ b Dconvenience þ it it it 1 2 negative feedback from customers. Our findings align with prior literature on the customer's asym- b Dsecurity þ b DCS PE þ b Dcomplexity it it it 3 4 5 metric reactions to service quality (Arbore and þ b Ddiscomfort þ b Dinsecurity þ b DCS NE þ e it it it it 6 7 8 Busacca 2009; Oh and Yi 2021). Further, we apply the first difference model to the ð3Þ bank subsample to examine what factors improved where, DPosR (DNegR Þ are changes in the ratio of it it or deterred the mobile banking app user experience positive (negative) ratings of bank i during week t. during the sample period. Estimation results (Table 5) imply that the improvements in Wells Fargo rat- 5. Results ing are mainly attributable to the increase in security, convenience, and easytouse. On the other hand, Cap- Table 4 shows the estimation results of the fixed ital One, which has the highest overall mean rating, effect model (level) and the first difference model experiences adverse reactions due to customer (change) for the weekly panel data. In the fixed-effect complaints regarding insecurity, CS_NE, discomfort, model, all constructs of mobile banking have signif- and complexity. The bank-level analysis shows that icant effects on the customer's overall rating (Rating) the quality dimensions of mobile banking apps have and the ratio of positive ratings (PosR) in the expected differential impacts on customer response from directions. In the negative ratio ratings (NegR) model, bank to bank. all explanatory variables show expected and 34 ASIA MARKETING JOURNAL 2022;23:28e37 Table 2. Text mining analysis of major quality dimensions in mobile banking. Construct Topic Terms Occurrence Content Sample Related Studies (%) Ease of use PE easy to use, simple, 66.15  The app is very easy to use. (Alalwan et al. straightforward,  This app gives me simplicity in 2016; Davis 1989; user-friendly, intuitive reviewing my account to check balances Gu, Lee, and Suh and pay off bills. 2009; Sharma 2019) NE hard, difficult, complicate 4.25  I am unsure why paying bills with the app got so difficult two weeks ago after the latest version or update. Most apps like this are so difficult and sensitive for security. Convenience PE convenient, track, notification, 26.02  Like how I can track and get payment (Benoit, Klose, and anytime, anywhere, fast, quick, alerts from text messages for approval Ettinger 2017; instant as to payment pending. Jebarajakirthy and Love the way I am instantly notified Shankar 2021; when a charge is posted to my account Shankar, Tiwari, by a merchant. and Gupta 2021) NE inconvenient, slow, crash, won't 18.17  A lot of crashes upon opening the app work, error, freeze, can't get, with several of these latest updates. white screen, please fix  I can't believe they still can't fix pay bills on the mobile app. Security PE security, secure 6.72  I feel totally secure using it to pay (Arcand et al. 2017; my payment. Chang and Chen I have touch id set up for the app, 2009; Shankar, which helps with security. Tiwari, and Gupta NE insecure, hacking, fraud, privacy, 1.06  I had fraudulent charges on my card. 2021; Sreejesh, concern  Constantly signing on today and it's Anusree, and making me leary if someone is Amarnath 2016) hacking my information. Customer PE {customer service, customer 3.61  Every time I talk to a customer service (Jun and Palacios Support support, staff, call} and {good, rep they are polite and very helpful. 2016; Shankar, great, excellent, best, helpful,  Customer service wait time is Tiwari, and Gupta friendly,polite} reasonable they are very friendly 2021) and helpful. NE {customer service, customer 0.81  Their customer service people are rude support, staff, call} and {poor, and don't know the first thing about worst, horrible, rude} true customer service. I've been a long time customer and have had the worst customer service over the years. Note: PE(NE) stands for positive(negative) user experience for the given construct. Table 3. Descriptive statistics and correlation matrix. DRating DPosR DNegR Deasytouse Dconvenience Dsecurity DCS_PE Dcomplexity Ddiscomfort Dinsecurity DCS_NE DRating 1.00 DPosR 0.91*** 1.00 DNegR 0.95*** 0.79*** 1.00 Deasytouse 0.30*** 0.30*** 0.30*** 1.00 Dconvenience 0.29*** 0.28*** 0.26*** 0.06 1.00 Dsecurity 0.31*** 0.28*** 0.30*** 0.06 0.05 1.00 DCS_PE 0.12*** 0.12** 0.10* 0.08 0.08 0.01 1.00 Dcomplexity 0.12*** 0.17*** 0.11** 0.20*** 0.14*** 0 0.06 1.00 Ddiscomfort 0.65*** 0.64*** 0.59*** 0.38*** 0.32*** 0.25*** 0.18*** 0.04 1.00 Dinsecurity 0.15*** 0.13** 0.15*** 0.09* 0.04 0.02 0.10* 0.01 0.12** 1.00 DCS_NE 0.12*** 0.11** 0.10* 0.10* 0.04 0.01 0.07 0.01 0.08* 0.01 1.00 Mean 0.004 0.001 0.001 0.0002 0.0004 0.0001 0.0004 0.0003 0.001 0.0002 0.0001 SD 0.44 0.12 0.12 0.1 0.09 0.05 0.04 0.05 0.12 0.03 0.03 Note: *p < 0.1; **p < 0.05; ***p < 0.01. ASIA MARKETING JOURNAL 2022;23:28e37 35 Table 4. Factors for improving and deterring mobile banking app user rating. Fixed Effect (Level) First Difference (Change) Rating PosR NegR DRating DPosR DNegR easytouse 0.50*** 0.12** 0.12** Deasytouse 0.10 0.01 0.06 (0.18) (0.05) (0.05) (0.15) (0.04) (0.04) convenience 0.91*** 0.21*** 0.21*** Dconvenience 0.45*** 0.09* 0.12** (0.21) (0.06) (0.06) (0.17) (0.05) (0.05) security 2.14*** 0.54*** 0.57*** Dsecurity 1.37*** 0.29*** 0.38*** (0.33) (0.10) (0.09) (0.27) (0.08) (0.08) CS_PE 0.77** 0.30*** 0.14 DCS_PE 0.53 0.12 0.10 (0.38) (0.11) (0.10) (0.34) (0.10) (0.10) complexity 2.56*** 0.85*** 0.58*** Dcomplexity 1.04*** 0.41*** 0.25*** (0.30) (0.09) (0.08) (0.26) (0.07) (0.08) discomfort 2.94*** 0.85*** 0.72*** Ddiscomfort 2.29*** 0.65*** 0.55*** (0.16) (0.05) (0.04) (0.15) (0.04) (0.04) insecurity 3.18*** 0.79*** 0.82*** Dinsecurity 3.33*** 0.87*** 0.84*** (0.54) (0.16) (0.15) (0.45) (0.13) (0.13) CS_NE 3.55*** 0.97*** 0.89*** DCS_NE 3.08*** 0.80*** 0.69*** (0.61) (0.18) (0.17) (0.50) (0.14) (0.15) Bank Fixed Effect Controlled Constant 0.004 0.001 0.001 (0.012) (0.004) (0.004) Obs. 569 569 569 565 565 565 R 0.80 0.78 0.77 0.55 0.53 0.47 Adj-R 0.80 0.78 0.76 0.54 0.52 0.46 Note: *p < 0.1; **p < 0.05; ***p < 0.01, Standard errors are in parentheses. Table 5. Factors for improving and deterring mobile banking app user rating by banks. Capital One Bank of America Chase Wells Fargo DRating DPosR DNegR DRating DRating DPosR DRating DPosR DNegR DRating DPosR DNegR Deasytouse 0.33 0.09 0.02 0.25 0.03 0.13 0.57 0.09 0.15 0.49* 0.03 0.21*** (0.24) (0.08) (0.07) (0.31) (0.09) (0.09) (0.35) (0.09) (0.11) (0.26) (0.08) (0.07) Dconvenience 0.53** 0.12 0.14* 1.04*** 0.30*** 0.22* 0.26 0.04 0.14 0.56** 0.06 0.21*** (0.24) (0.08) (0.07) (0.39) (0.11) (0.11) (0.37) (0.10) (0.11) (0.25) (0.07) (0.07) Dsecurity 0.55* 0.05 0.29*** 2.91*** 0.59*** 0.84*** 0.77 0.32** 0.05 0.79* 0.18 0.28** (0.33) (0.11) (0.10) (0.66) (0.19) (0.19) (0.60) (0.15) (0.18) (0.42) (0.12) (0.11) DCS_PE 0.37 0.06 0.14 1.48 0.32 0.29 0.22 0.17 0.05 0.55 0.15 0.16 (0.40) (0.14) (0.12) (0.95) (0.27) (0.28) (0.65) (0.17) (0.20) (0.63) (0.19) (0.17) Dcomplexity 2.07*** 0.63*** 0.49*** 0.52 0.24* 0.21 1.67*** 0.72*** 0.26 0.01 0.05 0.07 (0.39) (0.14) (0.12) (0.50) (0.14) (0.15) (0.56) (0.14) (0.17) (0.55) (0.16) (0.15) Ddiscomfort 2.80*** 0.79*** 0.72*** 2.03*** 0.62*** 0.44*** 2.20*** 0.52*** 0.58*** 2.36*** 0.70*** 0.56*** (0.22) (0.08) (0.07) (0.28) (0.08) (0.08) (0.36) (0.09) (0.11) (0.28) (0.08) (0.08) Dinsecurity 4.48*** 0.91*** 1.14*** 2.99*** 0.84*** 0.64** 3.32*** 0.90*** 0.93*** 3.63*** 1.16*** 0.79*** (0.59) (0.20) (0.18) (1.01) (0.29) (0.29) (0.88) (0.23) (0.27) (1.04) (0.31) (0.28) DCS_NE 4.23*** 0.96*** 1.12*** 2.68** 0.92*** 0.41 3.84*** 0.76*** 1.08*** 0.13 0.06 0.37 (0.69) (0.24) (0.21) (1.04) (0.30) (0.30) (0.99) (0.26) (0.30) (1.35) (0.40) (0.36) Constant 0.0003 0.0000 0.0002 0.004 0.0001 0.001 0.01 0.002 0.003 0.01 0.002 0.001 (0.01) (0.005) (0.004) (0.03) (0.01) (0.01) (0.03) (0.01) (0.01) (0.02) (0.01) (0.005) Obs. 136 136 136 143 143 143 143 143 143 143 143 143 R 0.78 0.68 0.73 0.63 0.61 0.53 0.39 0.43 0.29 0.66 0.60 0.67 Adj-R 0.77 0.66 0.71 0.60 0.59 0.51 0.35 0.40 0.25 0.64 0.58 0.65 Note: *p < 0.1; **p < 0.05; ***p < 0.01, Standard errors are in parentheses. do so, we collect 96,140 mobile app reviews for four 6. Conclusions U.S. banks: Bank of America, Capital One, Chase, 6.1. Summary and Wells Fargo. We first extract quality dimensions using the LDA topic model and interpret the factors This study examines the factors to improve as ease of use, convenience, security, and customer customer satisfaction in mobile banking services. To 36 ASIA MARKETING JOURNAL 2022;23:28e37 support based on the prior literature. We then (Shankar and Jebarajakirthy 2019). Thus, increasing conduct panel data analysis to investigate how user satisfaction with mobile banking apps is changes in each factor affect user ratings. Our esti- essential to retain loyal consumers and prevent mation results show that user ratings get improved customer churn. In particular, reducing complaints as positive responses to security and convenience due to poor customer support and technical errors is increase. However, user rating declines as reactions critical. Therefore, banks should utilize marketing to insecurity, negative experience with customer intelligence systems to monitor changing customer support, discomfort, and complexity accumulate. reactions reflected in user-generated reviews. Overall, we find that security is the most influential factor that affects user ratings. Moreover, analysis 6.4. Limitations and future research results for each bank suggest that the effect of each factor on customer satisfaction may differ from bank It is challenging to determine whether a phrase to bank. with negators means positive or negative until reading the actual review. Hence, this study con- 6.2. Theoretical implications structs and analyzes samples after removing phra- ses containing negators. However, the number of This study contributes to the existing literature in sample reviews will increase if advanced text min- the following two aspects. First, this study applies ing techniques correctly classify sentences with ne- text mining analysis to online reviews to explain gators. In such a case, the accuracy of detecting users' satisfaction with mobile banking services. As positive/negative perceptions of mobile banking the importance of mobile platforms increases, many service quality is also likely to increase. recent studies have investigated mobile app reviews This study does not examine banking-related to understand the motives of consumer behavior functions such as deposit, money transfer, bill pay- (Liu et al. 2019; Verkijika and Neneh 2021). How- ment, and credit score management. In contrast, this ever, only a few recent studies explore major quality study focuses on the four quality dimensions of dimensions of mobile banking app with customer- mobile banking service: ease of use, convenience, generated reviews (Leem and Eum 2021; Shankar, security, and customer support. With the merit of Tiwari, and Gupta 2021). Our study extends the rich consumer-generated text data, a future study prior research by identifying major quality di- can analyze how users' evaluation of each functional mensions using natural language processing and element affects mobile banking app satisfaction. machine learning methods. Research on customer response using online re- Second, this study is the first paper to study how views and text mining can provide in-depth insights the factors extracted through text analysis on online into consumer behavior towards financial services. reviews affect users' satisfaction. Existing studies related to mobile banking acceptance intention References primarily conduct a survey approach and identify influencing factors at a certain point in time. Unlike Alalwan, Ali Abdallah, Yogesh K. Dwivedi, Nripendra P.P. Rana, prior studies, we establish weekly panel data using and Michael D. Williams (2016), “Consumer Adoption of Mobile Banking in Jordan: Examining the Role of Usefulness, large-scale reviews to investigate the dynamic effect Ease of Use, Perceived Risk and Self-Efficacy,” Journal of En- of service quality factors on customer satisfaction. terprise Information Management, 29 (1), 118e39. Allenby, Greg (2012), “Keynote Speech at the Fall 2012 KMA Conference: Big Data 2.0,” Asia Marketing Journal, 14 (3), 1e5. 6.3. Managerial implications Arbore, Alessandro and Bruno Busacca (2009), “Customer Satis- faction and Dissatisfaction in Retail Banking: Exploring the The results of this study have the following im- Asymmetric Impact of Attribute Performances,” Journal of Retailing and Consumer Services, 16 (4), 271e80. plications. Security has the most significant influ- Arcand, Manon, Sandrine PromTep, Isabelle Brun, and ence on user satisfaction. In online reviews, words Lova Rajaobelina (2017), “Mobile Banking Service Quality and related to ease of use or convenience are left 4 to 10 Customer Relationships,” International Journal of Bank Market- ing, 35 (7), 1068e89. times more than words related to security. However, Benoit, Sabine, Sonja Klose, and Andreas Ettinger. (2017), user satisfaction with security has a greater impact “Linking Service Convenience to Satisfaction: Dimensions and on user ratings. These results suggest that banks Key Moderators,” Journal of Services Marketing, 31 (6), 527e38. Blei, David M., Andrew Y. Ng, and Michael I. Jordan (2003), need to pay special attention to security to increase “Latent Dirichlet Allocation,” Journal of Machine Learning customer satisfaction when designing mobile Research, 3, 993e1022. banking apps. Büschken, Joachim and Greg M. Allenby (2020), “Improving Text Analysis Using Sentence Conjunctions and Punctuation,” Currently, banks face increasingly fierce compe- Marketing Science, 39 (4), 727e42. tition to provide a better mobile services experience ASIA MARKETING JOURNAL 2022;23:28e37 37 Chang, Hsin Hsin and Su Wen Chen (2009), “Consumer Acceptance of Emerging Technologies: An Empirical Study of Perception of Interface Quality, Security, and Loyalty in Mobile Banking Services,” Decision Support Systems, 49 (2), Electronic Commerce,” Information & Management, 46 (7), 222e34. 411e7. Luo, Xueming, Christian Homburg, and Wieseke Jan (2010), Davis, Fred D. (1989), “Perceived Usefulness, Perceived Ease of “Customer Satisfaction, Analyst Stock Recommendations, and Use, and User Acceptance of Information Technology,” MIS Firm Value,” Journal of Marketing Research, 47 (6), 1041e58. Quarterly, 13 (3), 319e40. Mondres. (2020), “ABA Data Bank: Mobile Banking Adoption Fishbein, Martin and Icek Ajzen (1977), “Belief, Attitude, Inten- Accelerates,” ABA Banking Journal. Available at:https:// tion, and Behavior: An Introduction to Theory and Research,” bankingjournal.aba.com/2020/10/aba-data-bank-mobile- Philosophy and Rhetoric, 10 (2), 130e2. banking-adoption-accelerates/ . (accessed December 2, 2021). Gu, Ja-Chul, Sang-Chul Lee, and Yung-Ho Suh (2009), “De- Oh, Yun Kyung and Jisu Yi (2021), “Asymmetric Effect of Feature terminants of Behavioral Intention to Mobile Banking,” Expert Level Sentiment on Product Rating: An Application of Bigram Systems with Applications, 36 (9), 11605e16. Natural Language Processing (NLP) Analysis,” Internet Guo, Yue, Stuart J. Barnes, and Qiong Jia (2017), “Mining Research, https://doi.org/10.1108/INTR-11-2020-0649 Meaning from Online Ratings and Reviews: Tourist Satisfac- Proctor, Jasmine (2021), “Labour of Love: Fan Labour, BTS, and tion Analysis Using Latent Dirichlet Allocation,” Tourism South Korean Soft Power,” Asia Marketing Journal, 22 (4), Management, 59, 467e83. 79e101. Jebarajakirthy, Charles and Amit Shankar (2021), “Impact of Sampaio, Claudio Hoffmann, Wagner Junior Ladeira, and Online Convenience on Mobile Banking Adoption Intention: Fernando De Oliveira Santini (2017), “Apps for Mobile A Moderated Mediation Approach,” Journal of Retailing and Banking and Customer Satisfaction: A Cross-Cultural Study,” Consumer Services, 58, 102323. International Journal of Bank Marketing, 35 (7), 1133e53. Jeon, Jaihyun, Taewook Lim, Byung-Do Kim, and Junhee Seok Shaikh, Aijaz A. and Heikki Karjaluoto (2015), “Mobile Banking (2019), “Effect of Online Word of Mouth on Product Sales: Adoption: A Literature Review,” Telematics and Informatics,32 Focusing on Communication-Channel Characteristics,” Asia (1), 129e42. Marketing Journal, 21 (2), 73e98. Shankar, Amit and Charles Jebarajakirthy (2019), “The Influence Jun, Minjoon and Sergio Palacios (2016), “Examining the Key of E-Banking Service Quality on Customer Loyalty: A Dimensions of Mobile Banking Service Quality: An Explor- Moderated Mediation Approach,” International Journal of Bank atory Study,” International Journal of Bank Marketing, 34 (3), Marketing, 37 (5), 1119e42. 307e26. Shankar, Amit, Aviral Kumar Tiwari, and Manish Gupta (2021), Kekre, Sunder, Mayuram S. Krishnan, and Kannan Srinivasan “Sustainable Mobile Banking Application: A Text Mining (1995), “Drivers of Customer Satisfaction for Software Prod- Approach to Explore Critical Success Factors,” Journal of En- ucts: Implications for Design and Service Support,” Manage- terprise Information Management, https://doi.org/10.1108/JEIM- ment Science, 41 (9), 1456e70. 10-2020-0426 Krishnan, M.S., Venkatram Ramaswamy, Mary C. Meyer, and Sharma, Sujeet Kumar (2019), “Integrating Cognitive Antecedents Damien Paul (1999), “Customer Satisfaction for Financial into TAM to Explain Mobile Banking Behavioral Intention: A Services: The Role of Products, Services, and Information SEM-Neural Network Modeling,” Information Systems Frontiers, Technology,” Management Science, 45 (9), 1194e209. 21 (4), 815e27. Laukkanen, Tommi (2007), “Internet vs Mobile Banking: Sreejesh, S., M.R. Anusree, and Amarnath Mitra (2016), “Effect of Comparing Customer Value Perceptions,” Business Process Information Content and Form on Customers’ Attitude and Management Journal, 13 (6), 788e97. Transaction Intention in Mobile Banking: Moderating Role of Leem, Byung-Hak and Seong-Won Eum (2021), “Using Text Perceived Privacy Concern,” International Journal of Bank Mining to Measure Mobile Banking Service Quality,” Indus- Marketing, 34 (7), 1092e113. trial Management & Data Systems, 121 (5), 993e1007. Tirunillai, Seshadri and Gerard J. Tellis (2014), “Mining Market- Liu, Hongfei, Chanaka Jayawardhena, Sally Dibb, and ing Meaning from Online Chatter: Strategic Brand Analysis of Chatura Ranaweera (2019), “Examining the Trade-Off be- Big Data Using Latent Dirichlet Allocation,” Journal of Mar- tween Compensation and Promptness in eWOM-Triggered keting Research, 51 (4), 463e79. Service Recovery: A Restorative Justice Perspective,” Tourism Verkijika, Silas Formunyuy and Brownhilder Ngek Neneh (2021), Management, 75, 381e92. “Standing Up For or Against: A Text-Mining Study on the Luo, Xueming and Christian Homburg (2008), “Satisfaction, Recommendation of Mobile Payment Apps,” Journal of Complaint, and the Stock Value Gap,” Journal of Marketing,72 Retailing and Consumer Services, 63, 102743. (4), 29e43. Zhang, Jurui (2019), “What’s Yours is Mine: Exploring Customer Luo, Xin, Li Han, Jie Zhang, and J.P. Shim (2010), “Examining Voice on Airbnb Using Text-Mining Approaches,” Journal of Multi-Dimensional Trust and Multi-Faceted Risk in Initial Consumer Marketing, 36 (5), 655e65.

Journal

Asia Marketing Journal – Unpaywall

Published: Feb 4, 2022

There are no references for this article.