The Effect of Service Quality and Service Innovation on Customer Satisfaction at PT PLN (Persero) ULP Teluk Betung

This research uses quantitative methods, involving 100 respondents as samples. During this process, the instruments in the study were tested for validity and reliability. Followed by a classic assumption test consisting of normality test and multicollinearity test. Furthermore, hypothesis testing includes multiple linear regression analysis, coefficient of determination (R2 ), t test and f test. The purpose of this study was to determine whether service quality and service innovation affect customer satisfaction at PT PLN (Persero) ULP Teluk Betung. In this study, Service Quality affects Customer Satisfaction. This shows that the hypothesis is accepted. The value of Service Quality t count is 4.246> from t table 1.984. The results showed that Service Quality significantly affected Customer Satisfaction, Service Quality was rated positively in multiple regression analysis. The results of the study Service Innovation has a positive and significant effect on Customer Satisfaction, meaning that the hypothesis is accepted. The t test results show that the t value of Service Innovation is greater than the t table (7.256> 984). In multiple regression analysis, Service Innovation has a positive value on Customer Satisfaction. PLN is a state-claimed business that is delegated as a provider of electricity. Therefore, it needs a lot of innovation to attract new customers and service quality must be the main focus because it can create customer satisfaction and will affect customer loyalty.

This research uses quantitative methods, involving 100 respondents as samples.During this process, the instruments in the study were tested for validity and reliability.Followed by a classic assumption test consisting of normality test and multicollinearity test.Furthermore, hypothesis testing includes multiple linear regression analysis, coefficient of determination (R2 ), t test and f test.The purpose of this study was to determine whether service quality and service innovation affect customer satisfaction at PT PLN (Persero) ULP Teluk Betung.In this study, Service Quality affects Customer Satisfaction.This shows that the hypothesis is accepted.The value of Service Quality t count is 4.246> from t table 1.984.The results showed that Service Quality significantly affected Customer Satisfaction, Service Quality was rated positively in multiple regression analysis.The results of the study Service Innovation has a positive and significant effect on Customer Satisfaction, meaning that the hypothesis is accepted.The t test results show that the t value of Service Innovation is greater than the t table (7.256> 984).In multiple regression analysis, Service Innovation has a positive value on Customer Satisfaction.PLN is a state-claimed business that is delegated as a provider of electricity.Therefore, it needs a lot of innovation to attract new customers and service quality must be the main focus because it can create customer satisfaction and will affect customer loyalty.

INTRODUCTION
Competition in the business world is getting tighter, requiring company management to be more careful in determining its competitive strategy, in order to win the competition it faces.Company management is required to be able to design and implement marketing strategies that are able to create, maintain and improve service quality to customer satisfaction, which in turn can create high customer loyalty to the products offered.The company's success in providing quality service can be determined by the service quality approach that has been developed by Parasuraman, et al in Lupiyoadi (2013: 181).
PLN is a state-claimed business that is delegated as a provider of electricity.Based on the results of observations and interviews, there are several complaints at PT PLN (Persero) ULP Teluk Betung, namely soaring electricity bills, differences in meter records, as well as electricity tokens that fail input and blocked kwh meters.So there needs to be an improvement in the quality of these services.Furthermore, the lack of socialization about Many customers do not know about PLN's service innovations, such as the PLN Mobile application and other programs.This leads to a low level of usage of PLN's innovative services.Therefore, more innovation is needed to attract new customers and develop ties with existing customers.The Company strives to develop the availability of electricity for power system units, improve the network and bring updates to products and facilities applied across all service units.
In January 2008, PLN developed a data update-based assistance called prepaid electricity.Prepaid electricity has become an individual's number one decision to oversee electricity usage.The utilization of prepaid electricity benefits customers because with the service customers can regulate electricity usage and can set a budget for monthly electricity costs.In addition, customers are free from late disconnection penalties and automatically overcome complaints about inaccurate meter readings by officers.With this prepaid system, things that are not desirable will not happen again, so that between customers and PLN have a good relationship.In addition, people who use prepaid electricity can save more on electricity usage every day, because the control of electricity usage is entirely in the hands of the customer.
One of the factors that can affect customer satisfaction is service quality.Therefore, service quality must be the main focus of attention for PT PLN (Persero) ULP Teluk Betung because it can create customer satisfaction and will affect customer loyalty.Good or bad service quality of goods or services depends on the company's ability to consistently meet customer expectations.Satisfied customers will indirectly create loyalty, and encourage word of mouth recommendations, and can even improve the company's image in the eyes of customers.
According to previous research Prasetya et al. (2022) found that service quality and service innovation simultaneously and partially affect customer satisfaction.Rahman, S. (2019).Service quality and innovation simultaneously affect customer satisfaction at PT PLN (Persero) Manado area.Effendi, R, et al. (2019).There is a significant relationship between innovation and customer satisfaction.Risdah (2023) found that service quality and service innovation simultaneously and partially affect customer satisfaction.Mulyani & Yulianti (2023) found that service innovation and service quality simultaneously and partially affect customer satisfaction.This research is a hypothesis that states that the level of service quality of PT PLN (Persero) affects customer satisfaction in the Surabaya service area can be accepted.Saerang et al. (2018) From the research results obtained that service quality has a positive and significant effect on customer satisfaction.
Based on the description above, the authors draw conclusions with the formulation of the problem:

LITERATURE REVIEW Service Quality
According to Tjiptono (2015: 121) Service quality is a measure of how good the level of service provided is able to match customer expectations.Based on this definition, service quality can be realized through fulfilling customer needs and desires and the accuracy of their delivery to balance customer expectations.Parasuraman, et al. (in Kotler, 2016) states that there are five dimensions of service quality that affect customer satisfaction, namely tangibles (physical evidence), reliability, responsiveness, assurance, empathy.

Service Innovation
Owano et al., (2014) explain that service innovation is in the form of service products or service processes in the application of several technologies and systematic methods.Chen et al. ( 2016) suggest that service innovation has become an increasingly important consideration for companies, emphasizing innovative service initiatives, adoption and implementation of market concepts and value-added chains, with sustainable goals.Forms of service innovation can be in the form of mobile application development, online service development, new product and service development.

Customer Satisfaction
Customer satisfaction is the level of feeling happy or disappointed with the customer for the service received.Kotler in Cashmere (2017: 236) argues that customer decisions are an assessment of the customer's use of goods or services compared to expectations before their use.(Tjiptono 2014: 101) Customer satisfaction can be measured through several indicators, such as the level of customer perception of service quality, the level of customer loyalty, the level of willingness to recommend.

Framework of Thinking
The purpose of this study is to determine whether or not there is a relationship between the independent variables, namely service quality (x1) and service innovation (x2), and the dependent variable, namely customer satisfaction (y).The following is the framework used to formulate the hypothesis of this study:

METHODOLOGY
This type of research is quantitative research with the scope of discussing "The Effect of Service Quality and Service Innovation on Customer Satisfaction at PT PLN (Persero) ULP Teluk Betung".

Research Variables
This study uses three variables, namely: 1.Independent variables are some factors or signs that determine or influence whether other symptoms or factors exist or appear, such as service quality (x1) and innovation (x2).
2. The dependent variable is a group of signs or factors that exist or appear to be influenced or determined by the presence of certain independent variables, namely customer satisfaction (y).

Type and Source of Data
The data used in this study are as follows: 1.Primary data, which comes from data collected directly by researchers; this can also be referred to as "field research", because researchers will get information from the subjects to be studied directly in the field, namely customers of PT PLN (Persero) ULP Teluk Betung.2. Secondary data, which comes from data obtained from pre-existing sources.Such as journals, articles, websites, and books.

Population and Sample
According to Sugiyono (2014), population is a generalization area consisting of subjects or objects that have certain quantities and characteristics that have been determined by researchers to study and then draw conclusions.In this study, the population is all prepaid electricity customers of PT PLN (Persero) ULP Teluk Betung.Sample The author determines the sample members used in this research method are 100 respondents.In this study, the sampling method used was the "Simple Random Sampling" method, because all assumptions of the population were made randomly in other words the population was considered equal.

Data Collel ctil on Mel thods
A quel stil onnail rel , also callel d a poll, il s a rel sel arch tool consil stil ng of a sel ril el s of quel stil ons il ntel ndel d to collel ct data from thel pel rson bel il ng survel yel d.Thil s can bel donel pel rsonally vil a thel il ntel rnel t, tel lel phonel , computel r, or post (Mclel od, 2023)

Classil cal Assumptil on Tel st Normalil ty Tel st
Accordil ng to Ghozalil il n Alharil ts & Wahyono, (2022) thel purposel of thil s normalil ty tel st il s to tel st whel thel r thel confoundil ng or rel sil dual varil ablel s il n a rel grel ssil on model l havel a normal dil stril butil on.

Multil colonil el ril ty Tel st
Accordil ng to Ghozalil il n Putril & Susantil , (2022) Multil collil nel aril ty tel stil ng il s thel procel ss of analyzil ng how el xil stil ng rel grel ssil on el quatil ons rel latel to el ach othel r as il ndel pel ndel nt varil ablel s.To sel el thel prel sel ncel of multil collil nel aril ty il n thel rel grel ssil on el quatil on, il t can bel del tel rmil nel d by usil ng thel tolel rancel valuel and thel varil atil onal il nflatil on factor (VIl F).Il f thel VIl F valuel il s bel low or < 10 and thel Tolel rancel valuel il s abovel > 0.1, il t can bel concludel d that multil colonil el ril ty doel s not occur.

Hel tel roscel dastil cil ty Tel st
A condil til on whel rel thel rel il s an il nel qualil ty of varil ancel from thel rel sil duals for all obsel rvatil ons il n thel rel grel ssil on model l.How to tel st wil th thel Glel jsel r Tel st.Thel tel st il s donel by rel grel ssil ng thel il ndel pel ndel nt varil ablel s on thel absolutel rel sil dual valuel .Rel sil dual il s thel dil ffel rel ncel bel twel el n thel valuel of varil ablel Y and thel prel dil ctel d valuel of varil ablel Y, and absolutel il s thel absolutel valuel (all posil til vel valuel s).Il f thel sil gnil fil cancel valuel bel twel el n thel il ndel pel ndel nt varil ablel and thel absolutel rel sil dual> 0.05 thel n thel rel il s no hel tel roscel dastil cil ty.

Hypothel sil s Tel st Partil al Tel st (t Tel st)
To del tel rmil nel whel thel r thel il ndel pel ndel nt varil ablel (X) and thel del pel ndel nt varil ablel (Y) havel a partil al il nfluel ncel on el ach othel r, thel t tel st il s pel rformel d.Thel probabil lil ty valuel of thel il ndel pel ndel nt varil ablel il s usel d to del tel rmil nel il ts sil gnil fil cancel to thel del pel ndel nt varil ablel .Concurrel nt Sil gnil fil cancel Tel st (F Statil stil cal Tel st).Thel purposel of thel F tel st il s to il nil til atel a rel grel ssil on model l rel latil ng to thel il nfluel ncel of all il ndel pel ndel nt varil ablel s and thel del pel ndel nt varil ablel sil multanel ously.Thel rel sults arel mel asurel d wil th a sil gnil fil cancel lel vel l of 5% or 0.05.Thel F tel st il s conductel d as follows: 1. Del tel rmil nel thel formulatil on of thel null hypothel sil s and altel rnatil vel hypothel sil s: Ho: b = b = 0, mel anil ng thel rel il s no el ffel ct of X1. 2. Makel a del cil sil on basel d on thel F-count tel st.Il f thel probabil lil ty of thel F-count el rror ratel il s lel ss than 5%, Ho il s rel jel ctel d and Ha il s accel ptel d; othel rwil sel , il f thel probabil lil ty of thel F-count el rror ratel il s morel than 5%, Ho il s accel ptel d and Ha il s rel jel ctel d, whil ch il ndil catel s that thel il ndel pel ndel nt varil ablel s do not affel ct thel del pel ndel nt varil ablel sil multanel ously.

Rel lil abil lil ty Tel st
Thel rel lil abil lil ty tel st il s carril el d out to del tel rmil nel how consil stel nt thel mel asurel mel nt rel sults arel il f thel mel asurel mel nt il s carril el d out twil cel or morel wil th thel samel symptoms wil th thel samel mel asuril ng il nstrumel nt (Sil rel gar, 2015).Thel rel lil abil lil ty of thel rel sel arch il nstrumel nt wil ll bel tel stel d wil th Cronbach's Alpha.Thel cril tel ril on for thel rel lil abil lil ty of thel rel sel arch il nstrumel nt il s il f thel rel lil abil lil ty coel ffil cil el nt (r11) il s grel atel r than 0.70.Thel followil ng tablel shows thel tel st rel sults: A varil ablel can bel consil del rel d rel lil ablel il f il t has a Cronbach Alpha valuel of morel than 0.70 (Ghozalil , 2016).Basel d on thel rel lil abil lil ty tablel , thel Cronbach's alpha valuel s for thel Sel rvil cel Qualil ty (X1), Sel rvil cel Il nnovatil on (X2) and Customel r Satil sfactil on (Y) varil ablel s arel 0.905, 0.937, and 0.933 rel spel ctil vel ly.Thel rel forel , il t can bel concludel d that all il tel m statel mel nts arel rel lil ablel bel causel thel Cronbach's alpha valuel of all varil ablel s il s morel than 0.70.

Classil cal Assumptil on Tel st Hel tel roscel dastil cil ty Tel st
Basel d on thel output bel low, il t shows that thel Glel jsel r tel st valuel can bel sel el n that thel sil g> 0.05.For thel Sel rvil cel Qualil ty sil g valuel of 0.687, and thel sil g valuel on thel Sel rvil cel Il nnovatil on varil ablel il s 0.885.So il t can bel concludel d that thel rel arel no symptoms of hel tel roscel dastil cil ty.

Multil collil nel aril ty Tel st
From thel rel sults of data procel ssil ng bel low, il t il s obtail nel d that thel tolel rancel valuel il s grel atel r than 0.10 or thel samel as thel VIl F valuel il s smallel r than 10, so il t can bel concludel d that thel rel grel ssil on model l shows no multil collil nel aril ty.

Coel ffil cil el nt of del tel rmil natil on (R2)
By usil ng thel coel ffil cil el nt of del tel rmil natil on tel st, wel can del tel rmil nel thel el xtel nt to whil ch thel il ndel pel ndel nt varil ablel el xplail ns thel attachmel nt varil ablel .Thel coel ffil cil el nt of del tel rmil natil on for two il ndel pel ndel nt varil ablel s il s del tel rmil nel d by thel valuel :

Thel El ffel ct of Sel rvil cel Il nnovatil on on Customel r Satil sfactil on
Thel rel sults of thel study Sel rvil cel Il nnovatil on has a posil til vel and sil gnil fil cant el ffel ct on Customel r Satil sfactil on of PT PLN (Pel rsel ro) ULP Tel luk Bel tung, mel anil ng that thel hypothel sil s il s accel ptel d.Thel t tel st rel sults show that thel t valuel of Sel rvil cel Il nnovatil on il s grel atel r than thel t tablel (7,256> 984).Il n multil plel rel grel ssil on analysil s, Sel rvil cel Il nnovatil on has a posil til vel valuel on Customel r Satil sfactil on.
Thel rel sults showel d that thel sel rvil cel il nnovatil on varil ablel had a posil til vel and sil gnil fil cant el ffel ct on customel r satil sfactil on of PT PLN (Pel rsel ro) ULP Tel luk Bel tung.Thil s shows that sel rvil cel il nnovatil on il s an il ndil cator of customel r satil sfactil on.Bel causel nel w il nnovatil ons that can makel sel rvil cel el asil el r and morel convel nil el nt crel atel customel r satil sfactil on.Thel bel ttel r thel sel rvil cel il nnovatil on offel rel d, thel hil ghel r thel customel r satil sfactil on.

ADVANCEl D REl SEl ARCH
Thel suggel stil ons submil ttel d by thel author arel as follows: 1. PT PLN nel el ds to stril vel to il mprovel il ts sel rvil cel s, by provil dil ng sel rvil cel s to customel rs and thoroughly solvil ng all problel ms facel d by customel rs, so that customel rs fel el l satil sfil el d wil th thel sel rvil cel s provil del d. 2. Thel rel nel el ds to bel a good socil alil zatil on or promotil on stratel gy to thel publil c rel gardil ng thel advantagel s of el vel ry il nnovatil on madel by PT PLN, onel of whil ch il s prel pail d el lel ctril cil ty, both pel oplel who havel usel d thil s il nnovatil on and thosel who havel not so that el ducatil on can bel an optil on for thel communil ty il n el lel ctril cil ty usel rs.

Figure
Figure 1.Framework of Thinking Source: Rahman, S. (2019) Thel valil dil ty tel st il s carril el d out wil th a sil gnil fil cancel lel vel l of r tablel 5% (0.05).Thel valuel il s calculatel d basel d on thel correl latil on bel twel el n rel spondel nts' answel rs.Wil th del grel el s of frel el dom (df) = N -2 = 98, thel two-sil del d r tablel valuel il s obtail nel d at 0.1966, and thel r product momel nt tablel has an alpha sil gnil fil cancel of 5% (0.05).Thel il nstrumel nt il s del clarel d valil d il f r count il s grel atel r than r tablel and il nvalil d il f r count il s lowel r than r tablel .Basel d on thel tablel abovel , all statel mel nts for thel Sel rvil cel Qualil ty, Sel rvil cel Il nnovatil on, and Customel r Satil sfactil on varil ablel s arel del clarel d valil d bel causel r count il s grel atel r than r tablel 0.1966.

Tablel 1. Statel mel nt and Ratil ng Wel il ght
. Nel xt, a quel stil onnail rel il s crel atel d and il ts valil dil ty and crel dil bil lil ty arel tel stel d.Valil dil ty tel stil ng il s donel to el nsurel thel rel sel arch il nstrumel nt il s valil d or not.Valil d il ndil catel s that thel tool can bel usel d to mel asurel what should bel mel asurel d.Basel d on thel data abovel , thel il tel ms of el ach varil ablel gil vel n to rel spondel nts arel del clarel d valil d il f Rcount il s grel atel r than Rtabel l.Rel lil abil lil ty tel stil ng il s carril el d out to turn on thel abil lil ty of mel asuril ng il nstrumel nts to producel rel lil ablel or trustel d data.Cronbach Alpha was usel d to sprel ad thel rel lil abil lil ty of thil s study.Thel cril tel ril a for answel ril ng thel quel stil onnail rel basel d on thel assel ssmel nt arel as follows:

SEl ARCH REl SULT Rel spondel nts' Rel sponsel s Tablel 2. Frel quel ncy Dil stril butil on of Sel rvil cel Qualil ty (X1)
From thel tablel prel sel ntel d abovel , out of 100 rel spondel nts, il t can bel concludel d that 48 rel spondel nts or 48% statel d that thel Sel rvil cel Qualil ty was il n a vel ry good catel gory, 32% catel goril zel d il t as good, thel n 14% catel goril zel d il t as suffil cil el nt, thel n only 5% catel goril zel d il t as not good and 1% catel goril zel d il t as vel ry bad.So il t can bel concludel d that thel Qualil ty of Sel rvil cel il s il n thel vel ry good catel gory.

Frel quel ncy Dil stril butil on of Sel rvil cel Il nnovatil on (X2)
From thel tablel prel sel ntel d abovel , out of 100 rel spondel nts, il t can bel concludel d that 25 rel spondel nts or 25% statel d that Sel rvil cel Il nnovatil on was il n a vel ry good catel gory, 62% catel goril zel d il t as good, thel n 7% catel goril zel d il t as suffil cil el nt, thel n only 5% catel goril zel d il t as not good and 1% catel goril zel d il t as vel ry bad.So il t can bel concludel d that thel Sel rvil cel Il nnovatil on il s il n thel good catel gory.

Frel quel ncy Dil stril butil on of Customel r Satil sfactil on (Y)
From thel tablel prel sel ntel d abovel , out of 100 rel spondel nts, il t can bel concludel d that 31 rel spondel nts or 31% statel d Customel r Satil sfactil on (Y) il n thel vel ry good catel gory, 56% catel goril zel d il t as good, thel n 8% catel goril zel d il t as suffil cil el nt, thel n only 4% catel goril zel d il t as not good and 1% catel goril zel d il t as vel ry bad.So il t can bel concludel d that Customel r Satil sfactil on il s il n thel good catel gory.

dil ty Tel st
Thel followil ng arel thel rel sults of tel stil ng thel valil dil ty of thel rel sel arch varil ablel s:

Normalil ty Tel st Basel
d on thel output rel sults bel low, il t can bel sel el n that thel sil gnil fil cancel valuel il s 0.200> 0.05 so il t can bel concludel d that thel data tel stel d il s normally dil stril butel d.

Partil al Tel st
Thel el ffel ct of thel il ndel pel ndel nt varil ablel partil ally on thel del pel ndel nt varil ablel il s thel purposel of thel t tel st.Thel t tel st il s donel by comparil ng t count wil th t tablel or by lookil ng at thel sil gnil fil cancel column for el ach t count.Thel tel st rel sults arel shown il n thel followil ng tablel :

Tablel 11. Partil al T / Tel st Rel sults
Sourcel : SPSS output, 2024Basel d on Tablel 11, il t il s known that thel Sil g valuel of thel el ffel ct of Sel rvil cel Qualil ty (X1) on Customel r Satil sfactil on (Y) il s 0.00 <0.05 and thel t valuel il s 4.246> from thel t tablel 1.984 so il t can bel concludel d that H1 il s accel ptel d, whil ch mel ans that thel rel il s an il nfluel ncel bel twel el n Sel rvil cel Qualil ty (X1) on Customel r Satil sfactil on (Y).Sel cond Hypothel sil s Tel stil ng (H2) Il t il s known that thel Sil g valuel of thel el ffel ct of Sel rvil cel Il nnovatil on (X2) on Customel r Satil sfactil on (Y) il s 0.00 <0.05 and thel t valuel il s 7,256> t tablel of 1.984 so il t can bel concludel d that H2 il s accel ptel d, whil ch mel ans thel rel il s an il nfluel ncel on varil ablel Y.

Rel sults of thel Coel ffil cil el nt of Del tel rmil natil on (R²) Rel sults Model l Summary
Thel el ffel ct of thel il ndel pel ndel nt varil ablel (X) on thel del pel ndel nt varil ablel (Y) il s 80.1 pel rcel nt, as shown by thel rel sults of thel tablel abovel .Thel tablel also shows how thel Sel rvil cel Qualil ty and Sel rvil cel Il nnovatil on varil ablel s affel ct Customel r Satil sfactil on.Thel calculatil on rel sults show thel valuel of R = 0.890, and thel coel ffil cil el nt of del tel rmil natil on R Squarel il s 0.805, or 80.5%.Thel magnil tudel of thel coel ffil cil el nt of del tel rmil natil on mel ans that thel magnil tudel of changel il n thel 80.5% Customel r Satil sfactil on varil ablel il s il nfluel ncel d by Sel rvil cel Qualil ty and Sel rvil cel Il nnovatil on, thel rel mail nil ng 19.5% il s il nfluel ncel d by othel r varil ablel s not el xamil nel d il n thil s study.