UNDERSTANDING FEMALE SEGMENTS BASED ON BENEFIT OF LOYALTY PROGRAM

: Loyalty programs in banking need to look at the dynamics of female consumers, especially when digital businesses dominate marketing transactions. In the literature, segmentation studies are mostly carried out on retail services, airlines, and hotels, which are rare in banking, especially related to loyalty program services. Therefore, this study is expected to close the gap without a segmentation study in a banking context. This study aims to identify the female customer segment by assessing the bank's loyalty program and relating it to its perceived convenience, security, and reliability. The study employed the two most prominent banks in Indonesia, with 208 female customers as respondents. The purposive sampling method was used as a method of selecting samples. Data were reduced using factor analysis and categorized using cluster analysis. The main result identifies four factors underlying the benefits of loyalty programs: quality of communication, policy, rewards, and website quality. Three segments of loyalty program female consumers were identified: apathetic (25%), active (31%), and passive segments (44%). In further analysis, three segments of females were analyzed regarding the bank saving account's convenience, reliability, and security. Results confirm that all three segments were unique and distinguished one from another. This study's implication guides managing the types of female customers at the bank, especially loyalty programs.


INTRODUCTION
The urgency of this research is that loyalty programs in banking need to look at the dynamics of female consumers, especially when digital businesses dominate marketing transactions.Big data and technological change have made loyalty programs more common and complex (Stourm et al., 2020).Even though there is a potentially harmful effect on loyalty programs (Baker and Legendre, 2021), and this study is still fragmented, quite a lot has been studied (Kim et al., 2021).A loyalty program lets banks appreciate every customer who entrusts their funds to banks and actively transacts using various services.These services include internet banking, debit, and credit cards that customers shop at thousands of bank merchant partners.Some operational policies of bank loyalty programs: cover prizes that are not drawn; points are accumulated every month; customers are free to choose a variety of direct awards; the customer is not subject to the gift tax; the customers can exchange their points for other immediate prizes.Loyalty program for major banks in Indonesia has existed since 2013, and more than 13.9 million customers have enjoyed the benefits of this program.This loyalty program offers various prizes, including food and beverage, fashion, groceries, lifestyle and entertainment, gadgets and electronics, e-commerce, and transportation.
This study focuses on the female customer loyalty program in banking for two reasons.First, loyalty programs tend to be dominated by females, which is reflected in several facts; for example, the prevailing policy assumes that female consumers are more loyal than male consumers (Melnyk et al., 2009).They respond more positively to loyalty programs emphasizing personalization (Melnyk and van Osselaer, 2012) and innovation (Vilches-Montero et al., 2018).Moreover, their loyalty driver in malls consists of the atmosphere, physical design, and perceived quality of products and services (Haj-Salem et al., 2016).Their determinants of the perceived quality are trust and satisfaction (Abumalloh et al., 2020), and their experience with other guests at a hotel becomes a factor in loyalty (Khan et al., 2020).In addition, service quality significantly impacts female loyalty (Molinillo et al., 2021).Thus, it shows that female customers have unique characteristics compared to male cus-tomers.Second, loyalty program managers need to understand more deeply the characteristics of female customers in loyalty programs, mainly how the overall anatomy of customers interprets the loyalty programs offered.This understanding will lead to strategies for managing loyalty programs for female customers.The need to understand the female customers is paradoxical to the availability of knowledge regarding groups of female customers in loyalty programs.Literature studies related to loyalty programs are limited in elaborating on this interest.Several studies analyze aspects; for example, points are the most influential on female satisfaction on digital loyalty programs.The program that has the most influence on their loyalty is the e-coupon (Panjaitan, 2021).This study's gap is that previous loyalty program studies mainly focused on retail, airlines, and hotels and rarely on banking (Chen et al., 2021).Therefore, the expected managerial contribution from this study is to guide loyalty program managers in mapping and managing programs in the female segment according to their characteristics.In addition, this study adds to the knowledge of limited female segmentation references for literature.The novelty of this study is to map female customers based on the benefits sought in government banking, and the context of the bank loyalty programs is also reflected in this study.This study aims to segment female customers based on their perception of assessing the bank's loyalty program and relating it to their perception of its convenience, security, and reliability.

LITERATURE REVIEW Female Customer Segment
Female customers have several unique characteristics that marketers need to consider in marketing activities.Complaint behavior studies have shown that female customers are more capable of developing solid associations at high level of abstraction (Gruber et al., 2009).Also, they link desired behavior of providing employees with some values, tend to be more emotionally involved, and need time to calm down and relax when complaining.In retail loyalty, female customers are more loyal to individual store types than branch stores and are more influenced by satisfaction in interacting with store employees (Audrain and Vanhuele, 2016).

No Studies
Context and segmentation basis Female customer segments (Hong and Koh, 2002) Context: Apparel market, female in Korea.Basis: Benefit segmentation Brand-oriented, budget-oriented, and fashion-oriented.(Bakewell and Mitchell, 2003) Context: no specific product, Adult Female Generation Y consumers in the UK.Basis: Consumer-style inventory Recreational quality seekers, recreational discount seekers, trendsetting locals, shopping and fashion uninterested, and confusing time/money conserving.(Ko et al., 2007) Context: Fashion, female consumers of Korean, European, and USA.Basis: fashion lifestyle Information seekers, sensation seekers, utilitarian consumers, and conspicuous consumers (Hanzaee and Aghasibeig, 2010) context: No specific product, generation Y females in Iran Basis: the decision-making style.
Fashion consciousness, perfectionism, high-quality seekers, and price-value consciousness, (Hur et al., 2010) Context: Kitchen appliance market, female in the USA.Basis: lifestyle segmentation Wellbeing-oriented, social-and diningoriented, family-oriented, innovationand action-oriented, price-conscious, and convenience-oriented (Chan and Ng, 2012) Context: no specific product, female secondary school students in Hong Kong.Basis: gender roles and identities, ideal female images, and liking of global brands Middle of the roaders, achievers, conservatives, and inactive (Chan and Ng, 2013) Context: no specific product, adolescent girls in Mainland China Basis: psychographic segmentation Conformists, aggressive pursuers, image protectors, and single-handers.(Yıldırım et al., 2016) Context: Outlet center, female in Turkey.Basis: Consumer decision-making style.
Perfect-brand lovers, hedonist-fashion keepers, confused-impulsive buyers, price keepers.(Tsarenko and Lo, 2017) Context: Intimate apparel, females shopper in Australia Basis: consumer involvement Enthusiasts, dilettantes, and pragmatists (Milfelner et al., 2017) Context: Cosmetic surgery services, female in Slovenia Basis: attitudes toward the service Cluster 1, Cluster 2, Cluster 3, Cluster 4. (Amarjargal et al., 2018) Context: Luxury products, a female consumer in Mongolian Basis: Luxury value Passive shoppers, show-offs, rational value groups, and hedonists (Kartajaya et al., 2019) Context: Hijab, female in Indonesia Basis: Islamic fashion lifestyle Hijab factionist, aspirant Sharia oriented, moderate religious dressing, economical fashion follower, Sharia fashion follower, and pragmatic hijabers (Seebunruang, 2020) Context: Active sport tourists, females in Thailand Basis: Sport affinity Explicit active sport tourist, experimental active sport tourist, area active sport tourist, beginner Studies on female customers concerning loyalty programs can be traced from several studies.First, studies on the female shoppers (Chou et al., 2015) show that e-trust and e-satisfaction are associated with loyalty.Loyalty program innovation in a uniqueness that provides new experiences and can rarely be obtained is more attractive to female customer loyalty (Vilches-Montero et al., 2018).Women respond more positively to the loyalty programs that emphasize personalization, particularly personalization in personal settings (Melnyk and van Osselaer, 2012).In service loyalty, there is a positive relationship between customer involvement and the perceived value of women in forming loyalty (García-Fernández et al., 2020).Specifi-cally, in financial services, there is an awareness that needs and preferences are different in the women's market, affecting how women use and receive financial services (Jarden and Rappoldt, 2021).
Studies describing the segmentation of female customers show diverse dynamics.The context of previous researchers' attention was diverse but still focused on products specifically for women, for example, clothing or fashion (Hong and Koh, 2002;Ko et al., 2007;Tsarenko and Lo, 2017), kitchen equipment (Hur et al., 2010), cosmetics (Milfelner et al., 2017), and luxury products (Amarjargal et al., 2018).The basis of the segmentation is lifestyle, benefits, decision-making, attitudes, and psychographics.

Loyalty Program Segmentation
Loyalty programs are institutionalized incentive systems to improve consumer consumption over time, covering many program types (Kim et al., 2021).Studies on loyalty programs focus primarily on testing and applying theory in several contexts: social identity theory, social comparison theory, prospect theory, behavioral learning theory, social exchange theory, and the equity theory (Chen et al., 2021).Segmentation studies can be traced to several previous studies, where the studies conducted were still limited in number, context, and findings.Mapping of loyalty program customers is more often implemented in the context of retail services (Allaway et al., 2006(Allaway et al., , 2014;;Dogan et al., 2018;Thomas et al., 2006;Ieva and Ziliani, 2017;Kadir and Achyar, 2019), the hotel services (Tanford and Montgomery, 2015;Voohees et al., 2011).Some were banking (Mihova and Pavlov, 2018) and the mobile applications (Natalia et al., 2020).Table 2 shows that the basis used in loyalty programs segmentation varies from the base of the behavior, demographics, spending, attitudes, benefits, and other developments such as Recency, Frequency, and Monetary (RFM) analysis.
The findings of previous segmentation studies show two distinct patterns.The first pattern is a pattern that tends to be stratified in customer engagement in loyalty programs, namely from the lightest or low-intensity loyalty to high or high-intensity involvement.The findings of these studies refer to several names: iron/silver to platinum (Kadir and Achyar, 2019;Mihova and Pavlov, 2018); low to eco-true loyalty (Tanford and Malek, 2015); regular to advanced (Dogan et al., 2018); and very infrequent card shoppers to highly loyal shoppers (Allaway et al., 2006).The second pattern is a grouping based on a functional pattern, for example, special-treatment seekers, monetary-value seekers, and brand-advocates (Natalia et al., 2020); tilling and gardening (Thomas et al., 2006); and print lovers, online lovers, omni-media lovers, medium pickers, medium neutrals (Ieva and Ziliani, 2017).Therefore, the study of segmentation of loyalty programs is limited in the literature, and limited studies have been identified that specifically analyze segmentation of females in banking services.

METHOD
Research Design and Focus.A research de-sign is a strategy for gathering, measuring, and analyzing data that was developed to respond to research questions, where the study design is a survey from a strategy perspective, a cross-section in a time horizon, and an individual in a unit analysis perspective (Sekaran and Bougie, 2016).In addition, this research design is descriptive, namely a design that describes the characteristics or functions of a market and as the initial formulation before specific hypotheses are formulated (Malhotra, 2020).This study focuses on loyalty program services in banking in Indonesia.The unit of analysis is bank customers who were in the Surabaya while still be as customers and members of the loyalty program.

Population and Sample
The population in this study are female customers of bank funding products as owners of point rewards programs at the two largest banks in Indonesia.The reason for choosing these banks is that both banks are international banks and have excellent customer loyalty records based on Satisfaction, Loyalty, and Engagement (SLE) Awards held in Indonesia.Therefore, it is assumed that customers can express their perceptions about implementing the loyalty program at the bank.The sample size was determined by meeting the 10:1 ratio required for multivariate data analysis (Hair et al., 2018).
The sampling technique used was purposive sampling.Considerations for using this technique are its ability to draw logical generalizations (Saunders and Lewis, 2012) and limited access to obtain a sampling frame due to the confidentiality of information in banks.In addition, the sample criteria select participants because they have unique experiences, attitudes, or perception characteristics (Cooper and Schindler, 2014).The sampling criteria are customers with at least one savings account registered with the selected bank, who have at least six months of experience using rewards points, and who have used reward points in the past month.

Data Collection and Analysis
Surveys were used, where the advantages allow respondents to think about questions, faster implementation, and the ability to cover a wide geographic position (Cooper and Schindler, 2014).
Trained undergraduate students conducted the survey, in which fieldwork management follows a selection, training, supervision, validation, and evaluation process (Malhotra, 2015).The self-filled questionnaires were distributed to volunteer respondents in the Surabaya, East Java.The respondents were approached personally in places easily found by the respondents, such as homes, banks, shopping centers, office buildings, or campuses.All the respondents were informed about the nature of the study and were asked to answer the questionnaires.

Measurement
The measurement was developed by a questionnaire design process (Malhotra, 2015) to ensure that the research data collected would be feasible.This study adopted items that have been used previously from the literature.The items were validated and modified to fit the banking context.In an initial pretest, three individuals who had to redeem rewards points reviewed the initial questionnaire.Minor revisions were made.The revised questionnaire was tested in the initial study (n = 42) to test the quality of the survey instrument.The final survey questionnaire contains two parts.The first section contains screening questions and demographic information of respondents.The second part consists of items related to the construct of the study in which respondents were asked to respond.The loyalty program was measured using eighteen items adopted and modified from study of Omar and Musa (2011).Other constructs were also adopted from previous studies, convenience or ease to use (Jiang et al., 2016), security (Xie et al., 2017), and reliability (Jiang et al., 2016).The scale used is a seven-point scale ranging from ("strongly disagree" = 1) to "strongly agree" = 7).

Sample Description
Table 3 and 4 shows the number and the percentage of samples based on age, occupation, number of accounts, length of time as a customer, use of points during the last month, and savings facilities.Based on the table 3 and 4, the majority of respondents aged 20 to 29 years (57.7%),work as employees (27.4%), have one or two savings accounts (92.3%), have been customers for six to 23 months (57%), and have used points for once during the last month (73%).In addition, most of them used savings facilities, and a few used the mobile banking (29.7%), internet banking (24.5 %), and SMS banking (11.1).Thus, respondents were quite representative, reflecting the young and the beginning users of loyalty program in banking.

Descriptive Analysis
Table 5 shows the descriptive data of the statement items in the study.From the table, female customers' perception of the loyalty program is average at a positive or high level (5.81), with the most positive response is providing an easily accessible website (6.10).Likewise, female customers' reactions to aspects of convenience (6.11), security (6.02), and reliability (5.91) of savings in banks showed a positive or high level.

Factor Analysis
The loyalty program measurement consists of 22 items.Cronbach's alpha for this construct is 0.851, indicating good internal consistency.Factor ability was further tested by exploratory factor analysis.Several criteria were used to obtain factorial quality; it was observed that all items correlated at least 0.5 with at least one other item and showed reasonable factorability.Second, evaluating the adequacy of the sample size using the Kaiser-Meyer-Olkin (KMO) method showed a good level (0.789).Bartlett's test of sphericity showed significant results (χ 2 (153) = 1.177E3, p = 0.00).Communality shows a score above 0.5, confirming that each item has some variance from other items.Thus the factor analysis is considered to have been following these overall indicators.
Principal component analysis was used because the primary objective was to identify and calculate a composite score for the factors underlying the short version of the item that influence the online buying behavior of female consumers.The initial eigenvalues indicate that the first four or five factors explain 61% of each.The solutions for the factors were examined using the varimax rotation of the factor-loading matrix.The four-factor solution was preferred, which explained 55% of the variance.The reason was (a) there was theoretical support from previous studies (Omar and Musa, 2011); (b) the distribution of eigenvalues on the screen plot after passing the four factors (Figure 1) tended to be flat and decrease, and (c) there was an insufficient number of primary loadings and d) there was difficulty in interpreting the fifth and subsequent factors.The alpha coefficient indicating the internal consistency of each construct was evaluated.The alpha coefficient of each construct shows good indications, namely: 0.894 for factor 1 (5 items), 0.855 for factor 2 (5 items), 0.770 for factor 3 (4 items), and 0.838 for factor 4 (4 items).The four factors are properly labeled based on the content and substance of the item and theoretical justification.The labeling results produce names: communication quality (factor 1), policy (factor 2), gifts (factor 3), and website quality (factor 4).Loading factor and communality based on component analysis from factor analysis are presented in Table 6.
Communication quality consists of benefit statement items informing participating outlets, updating the website, providing timely feedback, and reminding the expiry date of points and vouchers.The item "inform participating outlets" has the highest loading (0.804).The policy factor includes program benefit items to provide clear prerequisites for the participation, easy-to-understand procedures, a precise method of calculating points, points received quickly, and sufficient time to redeem points.The item stated that "clear participation conditions" became the highest loading factor (0.698).The prizes factor consists of three things: the quality, branded, and attractive gifts, and the "branded gift" item has the highest loading factor (0.826).Finally, the website quality factor consists of the statement items providing helpful, reliable, informative, and easy access.The item "providing an informative website" scored the highest (0.757).

K-Mean Cluster Analysis.
Cluster analysis identified and categorized customers based on similarities in the benefits sought from bank loyalty programs.Since the sample consisted of hundreds of respondents, the k-means cluster analysis was employed on the data set after factor analysis.The k-cluster analysis method is sensitive to outliers (Saxena et al., 2013); therefore, the data set was first checked to remove outliers.There are three outliers data and they were eliminated using the Mahalanobis distance approach (p<0.001)(Hair et al., 2018).Based on the suggestions of a previous study (Jurowski and Reich, 2000), study context, and objectives, the analysis provides how many clusters would be used.Cluster analysis was conducted in an alternative number of two, three, and four clusters.Finally, we chose three groups considering the ease of labeling and substance of characteristics of each cluster.The results of cluster analysis are as in Table 7.
The figure 2 shows the pattern of responses from the female customers to the loyalty program, where there are significant differences between segments.Based on Figure 2, the three-cluster solution is Cluster 1 (25 percent members), Cluster 2 (31 percent members), and Cluster 3 (44 percent members).They are out of 208 respondents.Cluster 1 has the lowest score on communication quality (-0.476) and the highest on the website quality (0.433).Cluster 2 is the lowest in quality of communication (-0.476), and the highest is the quality of the website (0.745).While in cluster 3, the lowest is website quality (-0.683), and the highest is the communication quality (0.452).Table 5 also shows the difference in the mean factor between clusters, where all factors are significant (p<0.01).Significant differences existed between the groups, indicating the cluster's main characteristic based on the loyalty program assessment.Based on these main characteristics, each segment can be labeled, namely Cluster 1 with the label "apathetic segment," Cluster 2 with the title "active segment," and Cluster 3 with the label "passive segment".

Figure 2. Cluster Comparison
The segment characteristics can also be seen from their detailed assessment of each loyalty program component (Table 8).The characteristics of the three segments confirm the description of the segment characteristics as the previous description (Table 5).The apathetic segment is reflec-ted in the website quality and low in the other factors; the active segment is indicated by the great attention to rewards, policies, and website quality; and the passive segment is reflected as independent in communication and pays attention to the rewards.Characteristics of each segment can also be identified from their assessment of the convenience, security, and reliability of their savings account (Table 9).The apathetic segment perceives savings as a convenience service, even though the level of convenience is the lowest compared to the active and passive segments.The active segment gives the highest convenience rating.On the security of savings, all segments gave a secure rating.The apathetic and passive segments responded similarly; the highest response to security came from the active segment.Finally, the active segment assesses that savings are reliable, similar to the other segments regarding reliability.
The demographic characteristics of the segment can also be observed in several ways (Table 10).The table shows no significant differences between segments in terms of age, occupation, amount of savings held, and the length of time as a customer; however, there are differences in the types of savings most frequently used (Χ2=11,311; p < 0.05).The apathetic segment is balanced between the savings user from the Bank Mandiri and the Bank BRI, the active segment is more from the Bank BRI, and the passive comes from the Bank Mandiri.

DISCUSSION
The Benefit Sought Factors of Bank Loyalty Programs.
The benefits of bank loyalty programs can be grouped into four factors: communication quality, policies, prizes, and website quality.The communication quality consists of informing participating outlets, updating the website, reminding the expiry date of points or vouchers, and providing timely feedback.In other words, this factor shows how banks manage communication with customers by informing, reminding, and giving feedback on matters of concern for female bank customers.Thus, female customers will get what they need and want to know from the loyalty program.This factor is similar to the usefulness of the information in previous studies (Omar and Musa, 2011) The policy benefits the loyalty program by providing precise participation requirements, easy-to-understand procedures, a clear way of calculating points, the speed at which the points are received, and sufficient time to redeem points.These factors indicate that the policy aspect is essential for female customers.A specific policy for female customers will make understanding, calculating, and projecting the loyalty program's benefits easier.The finding of these factors in loyalty programs is consistent with previous studies, which also made policies a factor in loyalty programs (Omar and Musa, 2011) Rewards are benefit factor in quality, branded, and attractive gifts.The most important aspect of the loyalty program is giving to loyal customers.The more reliable they are, the more they get rewards.This benefit is a qualification sought by female customers in loyalty programs.This finding is consistent with previous findings, which made rewards a significant factor in loyalty programs (Omar and Musa, 2011).This factor is similar to the benefits and rewards offered (Alshurideh et al., 2020) and tangible rewards (Ma et al., 2018).
Website quality benefits a loyalty program by providing a functional, reliable, informative, and easily accessible website.Female customers pay attention to website information sources to obtain loyalty program information.The most important aspect is a website that is informative or describes the information that is easy to understand and as needed.This factor is similar to the communication factor in previous studies (Omar and Musa, 2011) and website service quality (Le et al., 2020;Wijaya et al., 2021).

The Female Segment of the Loyalty Program
This study aims to classify female customers of loyalty programs in banking services.The findings of this study identified three segments of female customers: apathetic, active, and passive.This grouping pattern is similar to a stratified way, i.e., according to the intensity of customer involvement.Moreover, this grouping pattern is consistent with a similar practice to previous studies (Allaway et al., 2006;Dogan et al., 2018;Kadir and Achyar, 2019;Mihova and Pavlov, 2018;Tanford and Malek, 2015).However, the three segments show differences in their main characteristics regarding their perception of the bank's loyalty program.
Apathetic segment.This segment represents 25 percent of the sample, or the smallest group compare to other segments.This segment's main characteristic is that they pay less attention to the gifts, policies, and the communication quality than different segments.They only pay attention to the website quality of the bank's loyalty program.In terms of savings as the primary service they obtain from banks, this segment considers savings accounts easy, safe, and reliable.They also rate convenience, security, and reliability as the lowest compared to other segments.This segment comes from female customers at Bank Mandiri and BRI on a balanced basis.
Active segment.This segment represents 31 percent of the sample, or the second largest, those seeking information to make conscious and deliberate decisions (Roos and Gustafsson, 2011).The main characteristic of this segment concerning the bank's loyalty program is that they are female customers who are most sensitive to website quality, gifts, and policies.However, this group of customers is the least concerned with the quality of communication.Thus, they are more active in seeking information through website information sources.For them, the gifts are the primary motivation, the website is a source of information, and specific policies are necessary.This segment of the female customers also values a savings account's convenience, security, and reliability the most.Therefore, their savings account is easy to use, safe, and reliable.This segment comes mostly from BRI customers.
Passive Segment.This segment represents 44 percent of the most significant sample.The passive segment is similar to customers who do not seek information and have fewer conscious reasons to make decisions (Roos and Gustafsson, 2011).The main characteristic is the most sensitivity or attention to the quality of the communication and the slightest attention to the quality of the website.Regarding loyalty programs, the bank's efforts to actively remind, inform, update information, and provide feedback are of utmost importance.Gifts for them are not the most important.Data from the website is also less important to them.This segment also assesses that their savings are easy to use, safe, and reliable.Most of this segment comes from Bank Mandiri customers.

RECOMMENDATIONS
The three segments identified in this study have implications for managing the loyalty program.Segment management is based on their characteristics and concern in loyalty program.First, loyalty program managers must look at the differences in the female customer segments to determine how to promote the loyalty program to each segment.Each segment has different characteristics and concerns, so managers need to adjust the marketing strategy based on it.
Second, marketing the bank's loyalty program to the apathetic female customer segment requires different promotions.Those who pay less attention to rewards, policies, and quality aspects of communication show less interest in loyalty programs, even though they use them.The less interest is because they have been primarily customers for over a year.The essential thing in this segment to maintain loyalty through loyalty programs is to encourage interest in the loyalty program.The loyalty programs must increase their interest through various important promotions, primarily through the media website.Their attention to website sources of information indicates that this media can effectively attract them to take advantage of loyalty programs.
Third, the active segment requires a different way in which rewards are the primary motivation and a policy and source of information.Managers of bank loyalty programs need to maintain this segment by managing quality, branded gifts, and attracting the attention of female customers.Retaining them is also about communicating policies to increase clarity for them.The website is one of the most critical factors in this segment, so information on the website related to loyalty programs needs to be developed.This segment also requires less communication regarding loyalty programs due to information sources and transparent policies that meet their needs.For the active segment, loyalty program managers need to increase customer engagement by considering personal and social activities and encouraging customers to interact with each other after use (Lee et al., 2018).
Fourth, the strategy of managing the passive segment is determined by what this segment pays attention to, the quality of communication.Since most loyalty program users are in this group, they need attention.They are passive and pay more attention to the bank's communication, gifts are not the most important, and the website is less important.Thus, the effort required is to change from passive to active segments.This effort is made by increasing the intensity of two-way communication and their involvement.Activating them will increase interest in the loyalty program and their loyalty.Maintaining contact with passive segments is essential to increase relationship knowledge (Roos and Gustafsson, 2011).
The limitations of this study are related to the use of purposive sampling; therefore, the findings cannot be generalized to loyalty programs in other countries.Further studies can be carried out using a probabilistic sampling method and in a more significant number of samples.In addition, the research needs to be replicated in various contexts and segments of shoppers.The loyalty program used in the selected banks is two state banks.The segmentation technique in this study uses K-Mean Cluster, and the further studies need to use other approaches, for example, an innovative alternative to mixture regression modeling (Kim and Lee, 2011).

CONCLUSIONS
This study aims to segment female bank customers on savings services based on the benefits sought in the loyalty program.This study identifies four beneficial factors of loyalty programs: communication quality, policy, rewards, and website quality.The findings confirm the findings in previous studies, and the four factors were the basis for determining the female customer segment.This study identifies three segments based on the benefits sought: apathetic, active, and passive.The three segments differ in evaluating the loyalty programs and savings services' safety, convenience, and reliability.In addition, prominent differences in bank accounts were used most often.

CITATION
Sutarso, Y., Sekarsari, L. A., Ilfitriah, A. M., and Martha, L. S. 2023.Understanding Female Segments Based on Benefit of Loyalty Program.Jurnal Aplikasi Manajemen, Volume 21, Issue 2, Pages 516-533.Malang: Universitas Brawijaya.DOI: http://dx.doi.org/10.21776/ub.jam.2023.021.02.19.I N D E X E D I N D O A J -D i r e c t o r y o f O p e n A c c e s s J o u r n a l s A C I -A S E A N C i t a t i o n I n d e x S I N T A -S c i e n c e a n d T e c h n o l o g y I n d e x D i m e n s i o n s G o o g l e S c h o l a r R e s e a c h G a t e G a r u d a I P I -I n d o n e s i a n P u b l i c a t i o n I n d e x I n d o n e s i a n O N E S e a r c h C O R R E S P O ND I N G A U T H O R Y u d i S u t a r s o F a c u l t y o f E c o n o m i c a n d B u s i n e s s , U n i v e r s i t a s H a y a m W u r u k P e r b a n a s , I n d o n e s i a

Figure 1 .
Figure 1.Eigenvalue with the Number of Components

Table 2 . Studies on Loyalty Program Customer Segmentation No Studies Context and Segmentation Base Segments findings
Context: Green hotel practices in the USA; Basis: attitudinal loyalty, behavioral loyalty, and green importance.Low loyalty, spurious loyalty, ecospurious loyalty, eco-latent loyalty, true loyalty, and eco-true loyalty.Context: Loyalty program on book commerce in Indonesia; Basis: RFM Analysis Iron, gold, platinum 10 (Natalia et al., 2020)Context: e-loyalty programs from mobile applications users in Indonesia; Basis: benefit segmentation Special-treatment seekers, monetaryvalue seekers, and brand advocates.

Table 5 . Variables, Items, Means, and Standard Deviations Variables and items Means Standard deviations
RL4 What was promised in Savings at the bank is permanently kept.5.75 0.91 Note: LP: Name of Loyalty Program