Jurnal Ekonomi dan Bisnis.
Vol.
13 No.
3 September 2024
P - ISSN
: 2503-4413
E - ISSN
: 2654-5837.
Hal 301 Ae 311 The Influence Of Service Failure Severity On Brand Forgiveness Study On Telkomsel Users Sarah Anjani Mulyawan Department of Management.
Padjadjaran University.
Indonesia Thomas Budhyawan Yudha Department of Management.
Padjadjaran University.
Indonesia Rita Komaladewi Department of Management.
Padjadjaran University.
Indonesia Umi Kaltum Department of Management.
Padjadjaran University.
Indonesia Penulis Korespondensi Sarah Anjani Mulyawan sarah20010@mail.
Article Info Article History :
Received 11 Jun - 2024 Accepted 08 Aug - 2024 Available Online 15 Sep Ae 2024 Abstract The aims of this study are: .
To determine the service failure severity from internet network issues according to Telkomsel .
To determine the influence of service failure severity from internet network issues on brand forgiveness of Telkomsel and .
To determine the influence of service failure severity from internet network issues on each brand forgiveness dimension of telkomsel users.
The study used an online survey to collect data.
The sample size for this study was 100 respondents.
Data analysis was conducted using mean analysis and PLS-SEM.
Research Findings: The study findings indicate that: .
the service failure severity from internet network issues, according to Telkomsel users, is categorized as high.
the service failure severity from internet network issues has a significant negative influence on the brand forgiveness of Telkomsel users.
the service failure severity from internet network issues has a significant negative influence on each brand forgiveness dimension .
ognitive, affective, behaviora.
of Telkomsel users.
This study simplifies the research conducted by Alnawas et al.
by focusing only on two main variables, service failure severity and brand forgiveness, in order to provide a more focused and in-depth understanding of the relationship between severity of service failure and brand forgiveness.
Keyword : service failure severity, brand forgiveness, telecommunication cellular.
INTRODUCTION
Service failure is something that every company wants to avoid.
This is because service failure represents a company failure to deliver a product or service (Roschk & Gelbrich, 2014:.
Although it is something that companies strive to avoid, in the service industry, service failure is common, frequently occurs, and often cannot be avoided (Mesquita et al.
, 2023:
To determine the severity of a service failure, it is possible to measure the level of service failure severity.
Service failure severity can vary depending on the issue at hand and the customers perception (Cho et , 2.
Therefore, service failure severity ranges from low to high.
Service failure severity can influence critical aspects, one of which is brand The influence of service failure severity on brand forgiveness is evidenced by research conducted by Alnawas et al.
in the hotel industry in the UK.
The research findings indicate that service failure severity has a significant negative impact on brand forgiveness (Alnawas et al.
, 2023: 1.
The findings regarding the significant negative influence of service failure severity on brand forgiveness need to be taken into consideration by companies, including PT Telekomunikasi Selular (Telkomse.
Based on the service quality achievement report for the period of Q1 - Q4 of 2023.
Telkomsel has achieved good service quality.
In line with this good service quality, in 2023.
Telkomsel attained a customer satisfaction index (CSI) of 57 out of 10 (Telkom Indonesia, 2024: .
Although Telkomsel has achieved good service quality and a high CSI score, there are still numerous news articles and complaints related to service failures at Telkomsel circulating in online media throughout 2023.
One of the service failures frequently complained about by Telkomsel users on online media is related to internet network Internet network issues refer to problems within the internet network that cause difficulties for users in accessing the internet (Lamberti, 2.
Telkomsel needs to pay attention to service failure severity, particularly for service failures that are frequently complained about, such as internet network issues.
This is because there have been previous research findings stating that service failure severity has a significant negative impact on brand However, it should be noted that these research findings cannot conclusively apply to cases of service failure severity arising from internet network issues and brand forgiveness among Telkomsel users, as the previous studies were conducted in different industries and countries.
Based on the above description, this study aims to determine the service failure severity from internet network issues according to Telkomsel users, to determine the influence of service failure severity from internet network issues on brand forgiveness of Telkomsel users, and to determine the influence of service failure severity from internet network issues on each brand forgiveness dimension of Telkomsel users.
LITERATURE
REVIEW
AND
HYPOTHESIS DEVELOPMENT
Service Failure Severity According to Kussusanti et al.
service failure severity is defined as Aua customerAos assessment of the level of problems that occur in a serviceAy.
In this study, researchers define service failure severity as the consumer assessment of the level of problems that occur in a service, particularly those related to internet network issues.
Service failure severity can vary depending on the issue at hand and the customers perception (Cho et al.
, 2.
The indicators used to measure service failure severity in this study refer to several previous studies, such as Alnawas et al.
Raza et al.
Liu & Li .
Salagrama et al.
Roy et .
, and Kussusanti et al.
Brand Forgiveness According to Christodoulides et al.
, brand forgiveness is defined as Authe consumer's cognitive, affective, and behavioral response to a brand's perceived wrongdoing, with the aim of maintaining a constructive relationship with the brandAy.
In this study, researchers define brand forgiveness as the cognitive, affective, and behavioral responses of consumers after experiencing wrongdoing from a brand, particularly wrongdoing related to internet network issues, which aim to maintaining good relations between consumers and brands.
Brand forgiveness consists of cognitive, (Christodoulides et al.
, 2021 : .
The cognitive dimension involves consumers evaluations and thoughts after having a (Christodoulides et al.
, 2021 : .
The affective dimension reflects consumers feelings of betrayal, disappointment, and loss of trust in the brand (Christodoulides et al.
, 2021 : .
The unforgiveness and is related to switching behaviour (Christodoulides et al.
, 2021 : .
The dimensions and indicators used to measure brand forgiveness in this study refer to Christodoulides et al.
RESEARCH METHODS
This study used a single cross sectional According to Nunan et al.
, a single cross sectional design is a research design that takes samples from the target population at one specific time and the information obtained from the sample is only one time.
In this study, the sample was taken during February 2024 and the information obtained from the sample was only one time.
The data collection technique used in this research is the survey method.
The survey in this study was conducted online using Google Form.
The population in this study are Telkomsel users who have experienced internet network issues.
In determining the minimum sample size, this study refers to Chin .
This study refers to Chin .
because this study uses PLS-SEM for verification analysis.
According to Chin .
, the minimum sample size for PLSSEM is recommended between 30 and 100.
According to Hair et al.
, there is no identification problem when using a small sample size.
However, to improve the accuracy of PLS-SEM estimates, a larger sample size can be used (Hair et al.
, 2.
For this reason, in this study the number of samples used was 100 samples.
This study employs judgmental sampling.
The judgments for sampling in this study are Telkomsel users who have experienced internet network issues in the last three The three-month period was chosen to avoid bias related to respondent memories (Chatterjee, 2.
Then, the study sample consists of 95% Telkomsel prepaid service users and 5% Telkomsel postpaid service The percentage of the sample adjusts the percentage of prepaid and postpaid service users in Telkomsel, which is 95.
of prepaid service users and 4.
7% of postpaid service users (Bestari, 2.
This study utilizes a Likert scale with five response categories, referenced from Nunan et al.
Descriptive analysis conducted in this study employs mean analysis with SPSS 25.
The mean values in this study are categorized into five categories, as described by Kusumah .
Verification analysis in this study is conducted using partial least squares path modeling (PLS-SEM) with SMART PLS 4.
Two measurements are conducted in PLS-SEM for second order and Relationship Between Service Failure Severity and Brand Forgiveness When the service failure severity is classified as low, consumers tend to consider the loss as trivial and ignore their negative emotions (Sengupta et al.
, 2.
Meanwhile, when the service failure severity is classified as high, consumers tend to perceive greater losses, give more negative evaluations (Sengupta et al.
, 2.
, and need to manage the negative emotions that arise (Roehm & Brady, 2.
Based on this, consumers who experience service failure with service failure severity classified as low tend to give more positive cognitive, affective, and behavioral responses to perceived errors from a brand.
Meanwhile, consumers who experience service failure with service failure severity classified as high tend to give more negative cognitive, affective, and behavioral responses to perceived wrongdoing from a brand.
Thus, it can be said that service failure severity can influence brand forgiveness.
The influence of service failure severity on brand forgiveness has been proven by research conducted by Alnawas et al.
This research was conducted on the hospitality industry in the UK.
The results showed that service failure severity has a significant negative effect on brand forgiveness (Alnawas et al.
, 2.
Hypothesis In this study, more specific hypotheses were formulated to determine the influence of service failure severity on each dimension of brand forgiveness.
The hypotheses in this study are as follows:
H1 : Service failure severity has a significant negative influence on brand forgiveness.
H2 : Service failure severity has a significant negative influence on the cognitive dimension of brand forgiveness.
H3 : Service failure severity has a significant negative influence on the affective dimension of brand forgiveness.
H4 : Service failure severity has a significant negative influence on the behavioral dimension of brand forgiveness.
first-order.
Different employed in measuring second-order (Hair Jr.
et al.
, 2021: .
In this study, the second-order measurement is performed using a disjoint two-stage approach.
The steps for second order and first-order measurement in this study are referred to by Yamin .
and Hair Jr.
et al.
with some adjustments.
Category Telkomsel service Have experienced internet network
issues in the last
three months RUSULT AND DISCUSSION
Result
Descriptive Characteristics of Respondents Table 1
Descriptive Characteristics of Respondents Category Gender
Age
Domicile Occupation Monthly Income
Group
Female Male
17 - 26
27 - 36
37 - 46
47 - 56
> 56
Aceh Banten Central Java East Java East Kalimantan North Sumatera Riau South Sumatera Special Capital Region of Jakarta Special Region of Yogyakarta West Java Civil Servant Entrepreneur Housewife Private Employee Students Teaching Assistant Unemployed < Rp1.
Rp1.
000 Rp2.
> Rp2.
- Rp4.
> Rp4.
- Rp5.
Frequency in internet network problems during the last three Percentage (N=.
Group > Rp5.
- Rp7.
Rp10.
Prepaid service Postpaid Percentage (N=.
Yes 1 - 2 times 3 - 4 times 5 - 6 times 7 - 8 times > 10 times The majority of respondents are female, aged 17 to 26, residing in West Java, students, with a monthly income ranging from less than Rp1,000,000 to Rp2,500,000, and users of Telkomsel prepaid service.
All respondents have experienced internet network issues in the last three months, with the majority experiencing such problems 3-4 times during this period.
Mean Analysis for Service Failure Severity Table 2 Mean for Service Failure Severity Indicator Items SFS.
In my opinion, the internet network issues I experienced are a major problem.
SFS.
In my opinion, the internet network issues I experienced have a significant SFS.
In my opinion, the internet network issues I experienced are an important SFS.
In my opinion, the internet network issues I experienced are a serious SFS.
In my opinion, the internet network issues I experienced are severe.
Mean Category Medium High High High Medium SFS.
SFS.
In my opinion, the internet network issues I experienced caused me In my opinion, the internet network issues I experienced made me angry.
Average High High High The average mean score for the indicators on brand forgiveness is 3.
A mean score of 33 is categorized as medium.
This indicates that, on average, respondents give a balanced cognitive, affective, and behavioral response to the experienced internet network issues.
one hand, the cognitive, affective, and behavioral responses of the respondents to the experienced internet network issues are However, on the other hand, the cognitive, affective, and behavioral responses of the respondents to the experienced internet network issues are positive.
The average mean score for the indicators on service failure severity is 3.
A mean score of 3.
43 is categorized as high.
This indicates that, on average, respondents rated the severity of internet network issues as high.
Internet network issues are categorized as high because, according to the average respondent, the experienced issues have a significant impact, considered important and serious, and cause discomfort and anger.
Second-Order Measurement Results using Disjoint Two Stage Approach First stage The result of model estimation linking the first-order construct, which is service failure severity, with the first-order components, which are cognitive, affective, and behavioral, is as follows:
Mean Analysis for Brand Forgiveness Table 3 Mean for Brand Forgiveness Indicator Items I think Telkomsel should get consequences for what happened.
I wish that others could see that Telkomsel is not I disapprove Telkomsel.
I feel sympathy for Telkomsel.
I have compassion for Telkomsel, which has wronged I feel that my faith in Telkomsel has been restored.
I avoid using Telkomsel.
I do not consider Telkomsel anymore when evaluating I am less likely to use Telkomsel Average * = Reverse scoring ** = Reverse question Mean Category Medium Medium Medium Medium Medium Medium Medium Medium High Medium Figure 1.
Model Estimation Result for First Stage of Disjoint Two Stage Approach The result of measurement model evaluation focused on the first-order component is as follows:
Table 4 Loading Factor.
Composite Reliability, and Average Variance Extracted for First Stage of Disjoint Two Stage Approach Cognitive Affective Loading Composite Average Variance Factor Reliability Extracted 0,876 0,703 0,833 0,892 0,787 0,844 0,644 0,871 0,764 A.
Behavioral 0,769 0,891 0,731 0,866 0,844 0,855 Table 6 Loading Factor.
Composite Reliability, and Average Variance Extracted for Second Stage of Disjoint Two Stage Approach Table 5 Fornell Larcker Criterion for First Stage of Disjoint Two Stage Approach Cognitive Affective Behavioral Cognitive Affective 0,839 0,639 0,803 0,538 0,627 The result of the measurement model evaluation is as follows:
Behavioral Service Failure Severity SFS.
SFS.
SFS.
SFS.
SFS.
SFS.
SFS.
Brand Forgiveness Cognitive Affective Behavioral 0,855 First-order components have loading factors Ou 0.
708 and composite reliability .
values ranging between 0.
80 - 0.
Therefore, the indicators of firstorder components can be considered Moreover, first-order components have average variance extracted (AVE) values Ou 0.
50 and the square root of AVE value is correlation values.
Thus, first-order components can be considered valid.
From the first stage of the disjoint two-stage approach, latent variable scores (LVS) for the first-order component are obtained.
These LVS will be used in the second stage of the disjoint two-stage approach Loading Factor Composite Average Reliability Variance Extracted Table 7 Heterotrait Monotrait Ratio for Second Stage of Disjoint Two Stage Approach Service Failure Severity Service Failure Severity Brand Forgiveness Second stage The result of model estimation linking the first-order construct, which is service failure severity, with the second-order construct, which is brand forgiveness, measured using LVS first-order cognitive, affective, and behavioral, is as follows:
Figure 2.
Model Estimation Result for Second Stage of Disjoint Two Stage Approach Brand Forgiveness First-order second-order construct have loading factors Ou 0.
708 and composite reliability .
values ranging 80 - 0.
Therefore, the indicators of first-order construct and second-order construct can be considered reliable.
Moreover, firstorder construct and second-order construct have average variance extracted (AVE) values Ou 0.
50 and heterotrait monotrait ratio (HTMT) values < 0.
Thus, first-order second-order construct can be considered valid.
The result of the structural model evaluation is as follows:
Table 8 Path Coefficient and R-square for Second Stage of Disjoint Two Stage Approach Path Coefficient Origin T Statistics Sampl (|O/STDE Valu e (O) V|) Service Failure Severity Ie Brand Forgivene Rsquar Original sample path coefficient values are negative, indicating that the influence of service failure severity on brand forgiveness is Moreover, the t-value is greater than the critical value .
and the p-value is smaller than the significance level .
, indicating that the influence of service failure severity on brand forgiveness is Thus, it is known that service failure severity has a significant negative influence on brand forgiveness.
Therefore.
H1 can be accepted.
It is also known that the Rsquare value for brand forgiveness is According to Chin .
, an R-square value of 0.
67 is considered high, an R-square value of 0.
33 is considered moderate, and an Rsquare value of 0.
19 is considered Therefore, the R-square value brand forgiveness is considered Figure 3.
Model Estimation Result for First-Order The result of the measurement model evaluation is as follows:
Table 9 Loading Factor.
Composite Reliability, and Average Variance Extracted for FirstOrder
A a a A A a a A AA a A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A A A a AA
A AA A
A AA A
A AA A
A a A A A AA A A AA A A AA A A AA AA AA a A A AA A A AA A A AA A A A AA AA A AA a A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A A A A A a A a a A AA A A A A A a a a a A A a A A AA A A A A A A AA
A AA A A A
A AA A A A
A AA A A A
A a A A A A A AA A A A A AA A A A A AA A A A A AA A A A A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A A AA A A A Table 10 Heterotrait Monotrait Ratio for FirstOrder Service CognitiveAffectiveBehavioral Failure Severity Service Failure Severity Cognitive Affective Behavioral First-Order Measurement Results The result of model estimation linking the first-order construct, which is service failure severity, with the first-order components, which are cognitive, affective, and behavioral, is as follows:
First-order construct and first-order components have loading factors Ou 0.
and composite reliability .
values ranging between 0.
80 - 0.
Therefore, the indicators of first-order construct and first-order components can be considered Moreover, first-order construct and first-order components have average variance extracted (AVE) values Ou 0.
and heterotrait monotrait ratio (HTMT) values < 0.
Thus, first-order constructs and first-order components can be considered valid.
The result of the structural model evaluation is as follows:
square values for cognitive, affective, and behavioral are considered moderate.
Discussion Service Failure Severity from Internet Network Issues According to Telkomsel Users The service failure severity from internet network issues, according to Telkomsel users, is categorized as high.
This is because the internet network issues are assessed to have significant impact, considered important and serious, and causing discomfort and anger.
The obstruction of various activities due to the internet network issues can be one of the possible reasons why Telkomsel users assess experienced internet network issues as having a significant impact.
The frequency of experienced internet network issues can also be one of the possible reasons why Telkomsel users assess the internet network issues as important and serious, and causing discomfort and anger.
This is because, based on the survey results from 100 respondents, it is known that over the past three months, 38% of respondents experienced internet network issues 3 - 4 times, 30% experienced internet network issues 5 - 6 times, 11% experienced internet network issues 7 - 8 times, and 1% experienced internet network issues more than 10 times.
Table 11 Path Coefficient and R-square for FirstOrder Service Failure Severity Ie Cognitiv Service Failure Severity Ie Affectiv Service Failure Severity Ie Behavior Path Coefficient Origin Statistics Sampl (|O/STDE Valu e (O) V|) Rsqua All original sample path coefficient values are negative, indicating that the influence of service failure severity on each dimension of brand forgiveness is Moreover, the t-value is greater than the critical value .
and the pvalue is smaller than the significance level .
, indicating that the influence of service failure severity on each dimension of brand forgiveness is significant.
Thus, it is known that service failure severity has a significant negative influence on each brand forgiveness dimension .
Therefore.
H2.
H3, and H4 can be It is also known that the R-square value for cognitive is 0.
362, for affective 314, and for behavioral is 0.
According to Chin .
, an R-square value of 0.
67 is considered high, an Rsquare value of 0.
33 is considered moderate, and an R-square value of 0.
is considered weak.
Therefore, the R- The Influence of Service Failure Severity from Internet Network Issues on Brand Forgiveness of Telkomsel Users The service failure severity from internet network issues has a significant negative influence on the brand forgiveness of Telkomsel users.
The possible reason why the service failure severity from internet network issues has a significant negative influence on the brand forgiveness of Telkomsel users is because when the service failure severity of internet network issues is classified as low.
Telkomsel users tend to consider the loss as trivial and ignore their negative emotions (Sengupta et al.
, 2.
Therefore, the cognitive, affective, and behavioral responses of Telkomsel users to the experienced internet network issues tend to be more positive.
Meanwhile, when the service failure severity of internet network issues is classified as high, consumers tend to perceive greater losses, give more negative evaluations (Sengupta et al.
, and need to manage the negative emotions that arise (Roehm & Brady, 2.
Therefore, the cognitive, affective, and behavioral responses of Telkomsel users to the experienced internet network issues tend to be more negative.
superiority over other mobile operators in five out of nine categories (Khatri, 2.
CONCLUSION
Based on the study results, it can be severity from internet according to Telkomsel users, is categorized as high.
The service failure severity from internet network issues has a significant negative influence on the brand forgiveness of Telkomsel users.
The service failure severity from internet network issues has a significant negative influence on each brand forgiveness dimension .
ognitive, affective, behaviora.
of Telkomsel users.
The service failure severity from internet network issues has the greatest significant negative influence on the brand forgiveness cognitive dimension of Telkomsel users and the smallest on the brand forgiveness behavioral dimension of Telkomsel users.
The Influence of Service Failure Severity from Internet Network Issues on Each Brand Forgiveness Dimension of Telkomsel Users The service failure severity from internet network issues has a significant negative influence on each brand forgiveness dimension .
ognitive, affective, behaviora.
of Telkomsel Based on the original sample of path coefficients, it is revealed that the service failure severity from internet network issues has the greatest significant negative influence on the brand forgiveness cognitive dimension of Telkomsel users and the smallest on the brand forgiveness behavioral dimension of Telkomsel users.
Cognitive dimension in brand forgiveness is related to consumers Meanwhile, behavioral dimension in brand forgiveness is related to consumers behavioral intentions.
Thus, the service failure severity from internet network issues has the greatest significant negative influence on the evaluation and thoughts of Telkomsel users and the smallest significant negative influence on the behavioral intentions of Telkomsel users.
The possible reason why the service failure severity from internet network issues has the greatest significant negative influence on the evaluation and thoughts of Telkomsel users is because Telkomsel users have the perception that Telkomsel is the best cellular This is not wrong because in 2023.
Telkomsel won the award as the best telecommunications operator in Indonesia (Haryanto, 2.
As a result of this perception, when service failures related to internet network issues occur, regardless of whether they are categorizing as high or low severity, the evaluation and thoughts of Telkomsel users following these issues become more sensitive.
The possible reason why the service failure severity from internet network issues has the smallest negative influence on behavioral intentions of Telkomsel users is because Telkomsel is considered a superior cellular operator compared to others.
This is known from Opensignal's mobile experience report in Telkomsel's REFERENCES