Operational Research in Engineering Sciences: Theory and Applications Vol. 4, Issue 3, 2021, pp. 1-20 ISSN: 2620-1607 eISSN: 2620-1747 DOI: https://doi.org/10.31181/oresta20403001s * Corresponding author. amalinashadrina@gmail.com (A. Shadrina), zulfa.fitri@mercubuana.ac.id (Z. F. Ikatrinasari) QUALITY IMPROVEMENT OF THE E-COMMERCE WEBSITE USING INTEGRATION OF KANO MODEL-IPA WITH QFD APPROACH Amalina Shadrina*, Zulfa Fitri Ikatrinasari Departement of Industrial Engineering, Mercu Buana University, Jakarta, Indonesia Received: 22 March 2021 Accepted: 07 June 2021 First online: 22 September 2021 Research paper Abstract: This study was conducted to analyze the quality of e-commerce websites, find out which items need improvement, and make improvement design e-commerce websites. This study uses the kano model and Importance Performance Analysis approach based on 7 dimensions with 39 attributes. This study used the survey results from 103 respondents who regularly use the e-commerce website Tokopedia.com. The results of the questionnaire through the validity and reliability tests were used to analyze the reliability of items and the feasibility of the results of the questionnaire. After that, the results of the questionnaire through the corellation and hypothesis tests were used to know the relationship between variables. The findings show that 7 attributes influence customer satisfaction. However, the findings are still nowhere near the expectations. Thus, it requires being the focus of improvement. Improvements in design of 7 attributes in the form of House of Quality with 6 technical features needed to improve the quality of e-commerce website Tokopedia.com. Therefore, effective management strategies can be applied to overcome the intense competition in the e- commerce industry. Key words: Website quality, Kano model, Importance Performance Analysis, E- commerce, QFD, Customer satisfaction 1. Introduction Technology has rapidly developed along with the times, which allows people to be able to work more easily and effectively (Wilson & Keni, 2018). One of the examples is the increase in internet development. The internet is a global network system that allows people to communicate globally, get information easily, as well as buy and sell products and services online. The internet has a big influence on how people work, shop, make payments, travel, and socializes (Kaur, 2011). The development of the internet in recent years has changed the way people do business for they are starting to do it in a digital manner rather than the traditional way. https://doi.org/10.31181/oresta20403001s Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 2 Technological advances and developments make life easier for human beings by making online business as well as buying and selling a trend in the future. In traditional trade, the seller and buyer have to meet each other in the same place, negotiate, and conduct transactions, which require both parties to agree on the prices. Meanwhile, in e-commerce or online trading, there is no need for buyers and sellers to meet each other. Instead, they meet through a website that acts as an intermediary that connects the two parties and facilitates the transaction. Therefore, buyers can buy products or services from sellers, while sellers can do business even if they do not have a place or shop (Wilson et al. 2019). E-commerce can be defined as a commercial transaction between two parties, organizations, and individuals which is carried out through a network or website (Psaila & Wagner, 2007). E- commerce covers several types of activities, such as retail shopping, banking, ordering food, ticketing, and others. Most people use the website as e-mail, looking for information, social media, and a place for online transactions. This makes buying and selling things online a trend in the future. Buying and selling online in the e- commerce industry is defined as the efforts made to market a product or service as well as to rebuild relationships between sellers and buyers through internet media (Kotler & Amstrong, 2014). Hence, a website quality assessment is needed based on several criteria and items to describe things expected of a website (Rondovic et al, 2017). Website quality can affect repurchase intentions. On e-commerce sites, repurchase intentions have been underlined by several researchers or actors (Wilson et al. 2019; Wilson & Keni 2018; Wilson 2018; Wilson & Christella 2019). In this case, the quality of website appearance is known as a factor that can build repurchase intentions to consumers. It is an advantage for the company when a consumer or customer has the intention to rebuy things from the same e-commerce company. There is a possibility that customers will buy products or services from other companies in the same industry. Moreover, some researchers have underlined that the relationship between customer loyalty and the success of e-commerce companies is the key to achieving success in the e-commerce industry since it can indirectly retain customers (Lee et al, 2009). One of the important things in repurchase in e-commerce is the website user interface features that are well designed that enable a positive impression on customers who will do a repurchase (Fan & Tsai, 2010). The quality of the website’s appearance has a considerable effect in influencing customer trust to shop on the site (Gregg & Walczak 2010). The quality of website design is related to the customer’s initial online buying behavior (Zhou et al., 2009). Online shopping websites are very important for businesses, retailers, and consumers where the features on the website are designed innovatively. For that reason, the e-commerce industry has to develop high-quality websites that provide a better online experience to attract and retain their customers in e-commerce (Stuart, 2003). The main challenges for e-commerce organizations are understanding the users' needs and developing them. E-commerce companies with websites that are difficult to use will protect a bad image on the Internet and weaken the company’s position in the e-commerce business (Barnes & Vidgen, 2002). This study aims to find out the extent of the satisfaction level of Tokopedia.com e-commerce web users through measurement between the current level of quality (perception) and the desired quality (expectations). The next goal is to be able to describe the position of the Tokopedia.com e-commerce website quality items to enable seeing which items are by the user's expectations and which ones need improvement. Therefore, it can be designed to propose improvements on the Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 3 Tokopedia.com e-commerce website. This study differs from previous studies. Previous studies only reached the stage of e-commerce website quality analysis, while this study proposed improvements on the results of e-commerce website quality analysis based on kano model and importance performance analysis methods using improvement scales Quality Function Deployment (QFD). Kang et al (2016) studied the evaluation of e-commerce websites based on the E-S-Qual method. A study conducted by Ilbahar and Cebi (2017) analyzed and classified design parameters according to customer expectations for evaluating the usefulness of e- commerce websites. Mohd and Zaaba (2019) argue on usability and security factor analysis on e-commerce websites. 2. Literature Review In general, quality is a characteristic of a product or service that reflects how well the product or service meets customer satisfaction (Negash et al. 2003). According to a study, customer perceptions on the quality of a website are based on features on the website that meet customer needs and impress the total excellence of the website. The previous researchers also mentioned that various dimensions of website quality which can be categorized as security, information quality, ease of use, and service quality. Customer perceptions of the quality of the website are based on features on the website that meet the needs and impressions of customers towards the website (Mona et al., 2013). Attractive website designs on e-commerce websites motivate consumers to engage in online shopping (Ganesh et al., 2010). According to some researchers, the features on the website have an important influence on online purchase intentions (Mansori et al., 2012). Another study argues that informative websites allow customers to compare and evaluate product alternatives thereby increasing customer satisfaction and thus influencing online purchase intentions (Hausman & Siekpe, 2009). The quality of information offered by a brand on online shopping sites is also an important factor. A specific study revealed that information quality has the highest influence on customer satisfaction among all dimensions of website quality (Kim & Jones, 2009). 2.1. Kano Model Kano model is used to determine how effective an indicator plays a role in improving service quality. The Kano’s attributes are divided into several categories. The first category is Must be (M) or basic needs, the customer simply accepts when it is fulfilled. However, if the product or service fails to satisfy the customer's basic needs, the customer will be very dissatisfied. For instance, although having unfriendly waiters causes customer dissatisfaction, having friendly service does not increase customer satisfaction since having a friendly waiters is a basic need (Garibay et al, 2010). The second category is One dimensional (O) or performance needs. The level of customer satisfaction is related to one-dimensional performance; thus, the higher the perceived service quality, the higher the customer satisfaction, and vice versa. When attributes are fulfilled, customers are satisfied; if they are not fulfilled, customers are dissatisfied. The level of customer satisfaction increases in accordance with the level of such attributes. Therefore, the categories of must be and one dimensional are conditions needed to achieve customer satisfaction (Basfirinci Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 4 et al, 2015). The third category is Attractive (A) or excitement needs, which shows a high level of customer satisfaction when fulfilled, but does not cause dissatisfaction when it is not fulfilled because it is not expected by the customer who may not know the product features. The fourth category is Reserve (R), which indicates that if an indicator in this category exists, the customer is dissatisfied, while if the opposite is true, the customer is highly satisfied. The fifth category is Indifferent (I), which indicates that the existence of indicators in this category seems to have no impact on customer satisfaction. The sixth category is Questionable (Q), which involves indicators that are still questionable since the possibility of customers being satisfied or dissatisfied is unclear (Dewi et al, 2018). The next step is determining the Kano’s category for each indicator. If (M + O + A)> (I + R), then the Kano’s category for the x- indicator is max {M, O, A), otherwise (M + O + A) <(I + R) then the Kano’s category for the x-indicator is max {I, R} (Kuo et al, 2012). The researcher will then use Importance Performance Analysis to process the attributes that fall into the M, O, and A categories. 2.2. Importance Performance Analysis (IPA) Importance Performance Analysis (IPA) has been one of the most extensively acknowledged systematic methods for measuring which items demand improvement. The research focused on the Importance Performance Analysis (IPA), Evans and Chon (1989) examined the capability of the IPA to control tourism strategies in two different places in the US. The investigation showed that local business workers were not content with the company’s performance. The next researcher used IPA to compare business competitiveness in Hong Kong with its main competitors in the Asia Pacific region. In analyzing the data, they used IPA for it can provide a basis for their business development strategies (Enright & Newton, 2004). Sorensson’s (2013) research results comparing national and international tourists using IPA revealed that national tourists place a higher level of importance on sustainable tourism than international tourists. Meanwhile, there are significant differences between national and international tourists in subsequent tourism. The results from the IPA consist of four quadrants. Quadrant I included the high importance but low performance. The items included in this quadrant represent the main items that need to be improved with top priority. Quadrant II involved high importance with a high level of performance. Thus, it does not need enhancement and the items recorded on this quadrant are meet expectations. Quadrant III is of low importance with low performance as well labeled as a low priority. Therefore, the items included in this quadrant are unimportant and do not pose a threat to the company. Quadrant IV means low importance with a high level of performance. Quadrant IV is labeled as possibly excessive. This shows items that are too much emphasized by the organization. Hence, organizations have to minimize these items. However, rather than focus in this quadrant, companies are obliged to allocate more resources to focus on items that are in quadrant I (Wong et al., 2011). 2.3. Quality Function Deployment (QFD) Quality Function Deployment is a way to make the needs and desires of customers as part of the design and production of a product or service. QFD is used by companies to identify customer needs in technical languages (Goetsch & Davis, 2014). QFD originated from Japan in 1972. It has been successful as a tool to help Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 5 systematic quality improvement teams translate market research and customer needs into technical characteristics to satisfy customer desires. In QFD, customer needs are reflected in the planning matrix or so-called 'quality home' or HOQ (Cohen, 1995). 3. Methodology This study is a survey study that used a quantitative approach by involving samples directly from the existing population. The purpose of this study was to determine the extent of user satisfaction on the quality of e-commerce Tokopedia.com websites, items that need to be improved and also suggestions for improvements on the quality of Tokopedia.com e-commerce websites. Variables measured to determine the quality of the website was to use seven dimensions including Website Design (WD), Product Quality (PQ), Security Quality (SQ), Delivery Quality (DQ), Delivery Accuracy (DA), Customer Service (CS), Customer Perception and Satisfaction (PS). Each dimension consisted of several question items as the basis for compiling the questionnaire. The sampling technique used in this study was simple random sampling where sampling was done randomly from the existing population. In general, based on the theory that sampling for factor analysis requires a minimum of 100 respondents (Kline, 1994). Thus, this study used a 90% confidence level with a 10% error margin where the sample size was 103 respondents. The e-commerce website used as a sample of this study was the Tokopedia.com e-commerce website. The data collection technique used was a questionnaire with a Likert scale and Kano scale. Likert scale consisting of 5 points, from strongly disagree (score 1) to strongly agree (score 5). Kano scale consisting of 5 points, that is M (Must be), O (One dimensional), A (Atractive), I (Indifferent), R (Reserve) dan Q (Questionable). After the questionnaire was distributed, validity and reliability tests were performed to determine whether the results of the questionnaire were suitable for analysis. An item that had a positive correlation with criteria (total score) and a high correlation showed that the item had high validity as well. Based on this, if the calculated r count value was smaller than the r table value, then then the question item became unvalid. Moreover, it was said to be valid if the calculated value was greater than the r table value. After testing the validity of the questions used in the study, the reliability test was then performed. The reliability test was carried out to find out whether the data collection tool showed the level of accuracy, stability or consistency. Thus, the data can be used for further analysis. An item is considered to be reliable if the value of Cronbach Alpha is bigger than the critical value. The specified critical value is 0.6. If the Alpha value is greater than 0.6 then it is reliable and if the Alpha value is less than 0.6 then it is not reliable (Sugiyono, 2016). After that, the results of the questionnaire through the corellation and hypothesis tests were used to know the relationship between variables. After the correlation and hypothesis testing, the data can be processed using the Kano model and also the importance performance analysis for further analysis. Variabel and attribut of the questionnaire is presented in Appendix 1. After the questionnaire data had been obtained, validity and reliability tests were performed to test the eligibility of the items and the accuracy of the questionnaire Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 6 results. After that, the results of the questionnaire through the corellation and hypothesis tests were used to know the relationship between variables. After being declared valid, reliable and passed hyphotesis, the result of questionnaire quality of the Tokopedia website was analyzed using kano model and Importance Performance Analysis. After getting the results from the kano model, then performed data processing Importance Performance Analysis which showed which indicators need improvement, the design of the improvement of the Tokopedia.com e-commerce website was improved by using the Quality Function Deployment. The design of the improvement was described in the House of Quality where there were several technical characteristics needed for improvement. 4. Results Based on the result of the questionnaire, the user of the Tokopedia.com website who became the respondents by gender involved in this study were 63% or 65 respondents were female, and the remaining 37% or 38 respondents were male. Meanwhile, seen from the age of respondents, 78% or 81 respondents were 20-30- year-old, 15% or 15 respondents were 31-40-year-old, and 7% or 7 respondents were 41-50-year-old. The types of work of the respondents were private employees 83% or 86 respondents, civil servants 9% or 10 respondents, students 4% or 4 respondents, and housewives 4% or 3 respondents. Based on the frequency of use of e-commerce websites, 19% or 20 respondents regularly used e-commerce website, 50% or 52 respondents frequently used e-commerce websites, and 31% or 31 respondents quite often used e-commerce websites. 4.1. Validity and Reliability Analysis The following are the results of the validity and reliability tests of the questionnaire results from 103 respondents involved in this study. Table 1. Validity Test Questions R Count Value R Table Value Decision R Count > R Table WD1 0.437 0.1638 Valid WD2 0.716 0.1638 Valid WD3 0.511 0.1638 Valid WD4 0.536 0.1638 Valid WD5 0.320 0.1638 Valid WD6 0.616 0.1638 Valid PQ1 0.637 0.1638 Valid PQ2 0.652 0.1638 Valid PQ3 0.726 0.1638 Valid PQ4 0.491 0.1638 Valid PQ5 0.272 0.1638 Valid SQ1 0.633 0.1638 Valid SQ2 0.579 0.1638 Valid SQ3 0.632 0.1638 Valid SQ4 0.524 0.1638 Valid SQ5 0.738 0.1638 Valid DQ1 0.615 0.1638 Valid DQ2 0.342 0.1638 Valid Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 7 Table 1. Validity Test (Continue) Questions R Count Value R Table Value Decision R Count > R Table DQ3 0.731 0.1638 Valid DQ4 0.726 0.1638 Valid DA5 0.387 0.1638 Valid DQ6 0.208 0.1638 Valid DQ7 0.487 0.1638 Valid DA1 0.544 0.1638 Valid DA2 0.632 0.1638 Valid DA3 0.404 0.1638 Valid DA4 0.685 0.1638 Valid DA5 0.466 0.1638 Valid CS1 0.251 0.1638 Valid CS2 0.210 0.1638 Valid CS3 0.265 0.1638 Valid CS4 0.251 0.1638 Valid CS5 0.716 0.1638 Valid CS6 0.670 0.1638 Valid PS1 0.618 0.1638 Valid PS2 0.526 0.1638 Valid PS3 0.620 0.1638 Valid PS4 0.684 0.1638 Valid PS5 0.562 0.1638 Valid Table 1 presents the results of the SPSS 25.0 calculation regarding the overall value of r count which is greater than the r table value of 0.1638 for 103 questionnaires. Consequently, it can be concluded that the whole question items in the questionnaire are valid. After that, a reliability test was made to measure the reliability of respondents' responses to the general items of the questions asked. According to Santoso (2010), the questionnaire is deemed to be reliable if the value of the Cronbach Alpha is above 0.60. The following are the results of the reliability test of 39 question items in the questionnaire. Table 2. Reliability Test Question Items Alpha Cronbach Description 39 0.931 Reliable Based on the calculation using SPSS 25.0 presented in Table 2 above, it can be inferred that the Cronbach Alpha coefficient value is 0.931. Thus, it can be concluded that the question items in the Tokopedia.com e-commerce website quality questionnaire had a good level of consistency. Therefore, the findings of this study can be accounted for and can be used for further data processing to provide solutions for improving the quality of e-commerce websites. 4.2. Corelation Analysis The relationship between each variable is referred to as correlation. Correlation refers to how a change in one variable causes a change in the direction of another Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 8 variable. The higher the correlation, the closer the absolute value is to one. As a result, "+" indicates a positive change direction, while "-" indicates a negative change direction. The following is a correlation test between variables where all the dimensions of the tested variables are correlated or have a relationship between variables, since the value of the correlation test results is <0.05 or there are ** and * signs appeared in Table 3. Table 3. Correlations Test WD PQ SQ DQ DA CS PS 1 WD 103 103 .680** 1 PQ 0 103 103 103 .529** .577** 1 SQ 0 0 103 103 103 103 .608** .705** .707** 1 DQ 0 0 0 103 103 103 103 103 .487** .588** .858** .683** 1 DA 0 0 0 0 103 103 103 103 103 .200* .228* 0.19 .233* .249* 1 CS 0.043 0.021 0.055 0.018 0.011 103 103 103 103 103 103 .739** .637** .748** .718** .743** .294** 1 PS 0 0 0 0 0 0.003 103 103 103 103 103 103 103 ** Correlation is significant at the 0.01 (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). 4.3. Hypothesis Analysis Hypothesis testing aims to test the effect of the quality of the Tokopedia.com website on customer satisfaction. The hypothesis was tested using regression analysis by determining seven sub-factors as independent variables, with customer satisfaction as the dependent variable. The regression analysis hypothesis test results, which are attached in Table 4, are as follows. Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 9 Table 4. Hypothesis Test Unstandardized Coefficients Standardized Coefficients T Sig. Adj R2 B Std. Error Beta (Constant) -2.156 1.141 -1.889 0.062 0.996 WD 1.193 0.056 0.226 21.435 0.000 PQ 1.034 0.060 0.173 17.128 0.000 SQ 0.933 0.078 0.160 12.029 0.000 DQ 1.144 0.055 0.225 20.968 0.000 DA 1.185 0.082 0.191 14.398 0.000 CS 1.235 0.071 0.115 17.329 0.000 PS 0.957 0.078 0.159 12.250 0.000 Hypothesis 1 predicts that the Website Design (WD) positively influences consumen satisfaction. The regression analysis of Hypothesis 1 has P=0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 1 is accepted. Hypothesis 2 predicts that the Product Quality (PQ) positively influences consumen satisfaction. The regression analysis of Hypothesis 2 has P=0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 2 is accepted. Hypothesis 3 predicts that the Security Quality (SQ) positively influences consumen satisfaction. The regression analysis of Hypothesis 3 has P=0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 3 is accepted. Hypothesis 4 predicts that the Delivery Quality (DQ) positively influences consumen satisfaction. The regression analysis of Hypothesis 4 has P=0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 4 is accepted. Hypothesis 5 predicts that the Delivery Accuracy (DA) positively influences consumen satisfaction. The regression analysis of Hypothesis 5 has P=0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 5 is accepted. Hypothesis 6 predicts that the Customer Service (CS) positively influences consumen satisfaction. The regression analysis of Hypothesis 6 has P= 0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 6 is accepted. Hypothesis 7 predicts that the Customer Perception and Satisfaction (PS) positively influences consumen satisfaction. The regression analysis of Hypothesis 7 has P=0.000<0.05, indicating a statistically significant correlation. Thus, Hypothesis 7 is accepted. 4.4. Kano Model and Importance Performance Analysis The Kano model was analyzed to determine the attributes that customers need as well as those that have the potential to become a source of innovation for the Tokopedia.com website. Each respondent's questionnaire was categorized into M (Must be), O (One dimensional), A (Attractive), I (Indifferent), R (Reserve), and Q (Questionable). After collecting 103 respondents, then the results of the number of each service attribute are calculated. Data processing in the Kano model for each attribute is determined by the following rules, namely if (M+O+A)>(I+R), then the service attribute category is max (O, A, M) and if (M+O+A)<(I+R), then the Kano’s category for the service attribute is max (I, R). Table 5 summarizes the results of the Kano model calculation. Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 10 Table 5. Kano Model No Code M O A I R Q M+O+A I+R Kano Category 1 WD1 61 29 12 0 0 1 102 0 M 2 WD2 73 18 10 2 0 0 101 2 M 3 WD3 41 7 52 1 0 2 100 1 A 4 WD4 37 45 21 0 0 0 103 0 O 5 WD5 42 50 9 1 0 1 101 1 O 6 WD6 28 59 16 0 0 0 103 0 O 7 PQ1 51 37 12 2 0 1 100 2 M 8 PQ2 59 33 11 0 0 0 103 0 M 9 PQ3 40 18 44 1 0 0 102 1 A 10 PQ4 32 31 39 0 0 1 102 0 A 11 PQ5 37 25 41 0 0 0 103 0 A 12 SQ1 46 53 4 0 0 0 103 0 O 13 SQ2 35 67 1 0 0 0 103 0 O 14 SQ3 31 48 15 2 0 7 94 2 O 15 SQ4 20 52 31 0 0 0 103 0 O 16 SQ5 48 41 6 4 1 3 95 5 M 17 DQ1 62 37 4 0 0 0 103 0 M 18 DQ2 59 12 30 1 0 1 101 1 M 19 DQ3 47 50 6 0 0 0 103 0 O 20 DQ4 36 21 39 3 0 4 96 3 A 21 DQ5 42 39 19 2 0 1 100 2 M 22 DQ6 38 55 10 0 0 0 103 0 O 23 DQ7 49 31 22 1 0 0 102 1 M 24 DA1 41 58 4 0 0 0 103 0 O 25 DA2 40 43 19 0 0 1 102 0 O 26 DA3 28 47 26 0 0 2 101 0 O 27 DA4 55 38 10 0 0 0 103 0 M 28 DA5 59 41 3 0 0 0 103 0 M 29 CS1 51 26 25 1 0 0 102 1 M 30 CS2 63 19 21 0 0 0 103 0 M 31 CS3 57 28 17 0 0 1 102 0 M 32 CS4 40 53 10 0 0 0 103 0 O 33 CS5 31 49 23 0 0 0 103 0 O 34 CS6 34 52 17 0 0 0 103 0 O 35 PS1 59 40 4 0 0 0 103 0 M 36 PS2 41 11 43 5 0 3 95 5 A 37 PS3 38 9 51 5 0 0 98 5 A 38 PS4 42 31 19 11 0 0 92 11 M 39 PS5 35 20 33 12 1 2 88 13 M According to Table 5, there are 17 attributes that are categorized as Must be (M), 15 attributes are categorized as One dimensional (O), and 7 attributes are categorized as Attractive (A) which means that all attributes are included in the categories M, O, and A which are tested, affect consumer satisfaction of e-commerce website users, while there are no attributes that fall into category I or R, which means that there is not a single attribute that does not affect customer satisfaction. Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 11 Therefore, in the next IPA processing, all attributes will be analyzed. The results of the IPA analysis, as shown in Figure 1, are as follows. Figure 1. Importance-Performance Analysis The items included in quadrant I were item WD2 with indicator the information on the website is effective, item PQ3 with indicator all products on the website are available, item DQ4 with indicator the websites offers discount or free shipping, item DA4 with indicator delivers products in accordance with the set conditions, item CS4 with indicator it provides me with convenient options for returning items, item CS5 with indicator this site handles products returns well, and item CS6 with indicator return policy is simple. After analyzing the questionnaire using the kano model and Importance-Performance Analysis (IPA) so that it is known which items need to be repaired (quadrant I), an improvement design using Quality Function Deployment (QFD) by arranging House of Quality. 4.5. Quality Function Deployment The first phase in preparing QFD was the matrix of consumer needs or Voice of Customer which was a list of items that were important to consumers. In this study, the results of Importance Performance Analysis (IPA) were used in determining the Voice of Customer by placing the items in quadrant I like the focus of improvement. Phase 2 was determining the ratio of improvement of each quality indicator on the Tokopedia.com website. To achieve desirable and measurable results, targets were measured as well. The following is the formula for calculating the value of the improvement ratio. (1) Quadrant I Quadrant III Quadrant IV Quadrant II Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 12 Using this formula, the determination of the improvement ratio was done by comparing the expected value with the perception obtained from the results of the questionnaire. The calculation of the improvement ratio in quadrant I items was as follows. Table 6. Improvement Ratio Code Indicator Expectation Perception Improvement Ratio (IR) WD2 The information on the website is effective 4.73 3.78 1.25 PQ3 All products on the website are available 4.76 3.79 1.26 DQ4 The website offers discount or free shipping 4.75 3.79 1.25 DA4 Delivers products in accordance with the set conditions 4.78 3.81 1.25 CS4 It provides me with convenient options for returning items 4.72 3.93 1.20 CS5 This site handles product returns well 4.81 3.78 1.27 CS6 Return policy is simple 4.66 3.70 1.26 The third phase was determining the technical characteristics. Technical characteristics are the response given by the company to user desires that need to be revised. Determination of the technical characteristics was conducted by benchmarking, discussion, and interviews with interested parties. Determination of technical characteristics included various features of the website, standards for completeness of information, consistency in implementing SOPs, refund policies, Labor qualifications, and availability of assurance products. The fourth phase was the benchmarking stage which was conducted by comparing Tokopedia.com e- commerce with similar companies. The fifth phase was the relationship analysis of what’s and How’s Matrixes. In this section, the researchers described the relationship between customer needs and the technical characteristics needed to meet customer needs. This section was marked with number 3 which means it had a positive correlation and number 9 which means strong positive. The analysis of the relationship between what and how matric is presented in Appendix 2. The sixth phase was determining the correlation of technical characteristics drawn on the roof of the House of Quality. Characteristic correlations were described by four symbols, including the black circle symbol that represented a strong positive relationship and the white symbol that represents a positive relationship. Additionally, a negative relationship was depicted by a black triangle symbol which means strong negative and white triangle which means negative. Further, the seventh phase was setting goals or targets for each technical characteristic which means the steps or strategies needed for the organization to achieve the specified technical features. The following were the targets or stages that have to be carried out to achieve the required technical characteristics. Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 13 Table 7. Target or Limit Value Technical Characteristic Target Variety of Website Features There are features about product info, shipping refunds, payment, and expedition tracking features Completeness of Information Providing direction to the seller in the procedure of providing information in marketing their product Consistency in SOP Implementation Providing strict standards and consequences for parties related to customer sevice or the seller in carrying out duties or regulations in selling products Refund Policy The website provides a good refund policy Labor Qualification Providing training to customer service websites Assurance Availability Providing a guarantee of products purchased by customers if something goes wrong Next, the final phase was determining the priority level. The priority level was used to determine which Target or Limit Value had the highest priority level and the lowest priority level. Technical Priorities is the result of multiplying the values contained in the technical characteristics with the value of importance to customer. The results of the technical characteristic values can be used to calculate the percentage value for total priorities. The greater the value, the greater the priority for improvement. The results of the House of Quality (HOQ) design to improve the quality of e-commerce websites, in the Tokopedia Company is presented in Appendix 3. 5. Discussions Value of technical priorities is obtained by multiplication between the values of the technical characteristics with the value of importance to customer. Example for calculate Technical Priorities of Variety of Websites Features =(9×5)+(9×5)+(9×5)+ (9×5)+(9×5)+(9×5)+(9×5)=315. Example for calculate Percent Total Priorities of Variety of Websites Features=(315/(315+315+315+285+225+285)=18%. Others Technical Priorities and Percent Total Priorities can be calculated as above formula. Based on Appendix 3, determining the top priority in improving website quality based on the value technical priorities became the technical focus of improving the quality of e-commerce websites that have to be done immediately. The technical characteristic with the highest weight was variety of website features, completeness of information and consistency in the implementation of SOP. A variety of website features is various features found on a website designed for marketing needs. Thus, the intended website can meet the desires of the user. Variety of website features where the website must provides features that can accommodate all information related to the product with all its policies. This includes features of the information that is accurate, reliable, timely, and also detailed. The website must provide tracking features for product shipments as well as by phone interaction with the customer service website. The standard of completeness of information in which the seller is most responsible for these characteristics. The standard of completeness of information including websites must provides accurate, reliable, timely, and also detailed information. Additionally, the website has to provide information on Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 14 product return policies, information on compensation in the event of an internal problem, and also information about the customer service website telephone number. In case SOP implementation, SOP must was made with strict standards. Moreover, there were consequences if the SOP was not carried out according to the standard. The strict standard was to give obligations to related parties both customer service and seller in complying with the rules in the applicable SOP. SOP on customer service including the ability of customer service website as a good mediator in the event of the return of goods between the seller and buyer, the ability of customer service in resolving the problem immediately, the ability of customer service in providing a solution in the form of compensation in the event of an internal problem, and interacting directly with the users. Meanwhile, according to the SOP, the seller is required to provide accurate, reliable, up to date, and detailed information. The second-highest weighted technical characteristic was refund policy and assurance availability. Refund policy was in effect for the product received which was not following the expectation of the buyer, such as a product in a defective condition, inadequate quantity, wrong color, and so on. The indicators included in the refund policy were product returns according to applicable regulations, immediately handled by the customer service, the existence of compensation, and the ease of communication between the user and the customer service website. The indicator required to be reviewed to prevent refunds was the website. In this case, the seller has to provide accurate and reliable information to minimize product returns. An assurance on the website is a guarantee given by the Tokopedia.com website when things go wrong. The indicators included in this regard are the website handling product returns well, as well as being able to provide compensation in the event of an internal problem. Indicators that also need to be reviewed were if the website does not provide accurate, reliable, and timely information, in addition to that if the customer service does not resolve the problem immediately and is also difficult to contact. The third-highest weighted technical characteristic was labor qualifications. Employees were the spearhead of the website quality delivery system. For employees to be able to meet the expectations of website users effectively, it requires support from the main management functions. This support can be in the form of equipment, information, and training in service standards. Indicators included in terms of workforce qualifications were handling product returns properly, addressing problems immediately, and also the ease of communicating with customer service websites wither through chat or by phone features. 6. Conclusion Based on the results of the integration of the Kano and IPA models, it has been proven effective in knowing which attributes need to be improved in enhancing the quality of the e-commerce website Tokopedia.com. Furthermore, there are seven attributes that need to be improved out of the 39 attribute studied. The Kano and IPA models' results for these seven attributes have a significant effect on customer satisfaction, but they are not as expected. Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 15 The seven attributes are the information on the website is effective, all products on the website are available, the website offers discount or free shipping, delivers products in accordance with the set conditions, it provides me with convenient options for returning items, this site handles product return well, and return policy is simple. The seven attributes that do not meet customer expectations are then evaluated with QFD, which results in a technical analysis that needs to be improved, including a variety of website quality, standard of completeness of information, consistency of SOP implementation, refund policy, workforce qualifications, and assurance availability. The integration of the Kano and IPA models can be used for future research with a larger sample size. Furthermore, additional research must be capable of analyzing two or more related e-commerce websites. Reference Anupriya Kaur Y. Medury, (2011), "Impact of the internet on teenagers' influence on family purchases", Young Consumers, 12(1),27–38 http://dx.doi.org/10.1108/17473611111114768 Barnes, S., & Vidgen, R. (2002). An Integrative Approach to the Assessment of E- Commerce Quality. Journal of Electronic Commerce Research, 3(3), 114-127. Basfirinci, C., & Mitra, A. (2015). A cross cultural investigation of airlines service quality through integration of Servqual and the Kano model. Journal of Air Transport Management, 42, 239–248. https://doi.org/10.1016/j.jairtraman.2014.11.005 bt Mohd, N. A., & Zaaba, Z. F. (2019). A Review of Usability and Security Evaluation Model of Ecommerce Website. Procedia Computer Science, 161, 1199–1205. https://doi.org/10.1016/j.procs.2019.11.233 Blut, M. (2016). E-Service Quality: Development of a Hierarchical Model. Journal of Retailing, 92(4), 500–517. https://doi.org/10.1016/j.jretai.2016.09.002 Cohen, L. (1995). Quality function deployment-how to make QFD work for you. MA, UK: Addison-Wesley Publishing. D. Kang, W. Jang, Y. Park. (2016). Evaluation of e-commerce websites using fuzzy hierarchical TOPSIS based on E-S-QUAL, Applied Soft Computing Journal, http://dx.doi.org/10.1016/j.asoc.2016.01.017 Dewi, A. R., Putri, C., and Winarko. (2018). “Peningkatan Kualitas Jasa Fasilitas Kesehatan dengan Integrasi Metode IPA dan KANO”, Jurnal Sistem dan Management Industri., 2(2), 67-76. https://doi.org/10.30656/jsmi.v2i2.835 Enright, M. J., & Newton, J. (2004). Tourism destination competitiveness: A quantitative approach. Tourism Management, 25, 777–788. https://doi.org/10.1016/j.tourman.2004.06.008 Evans, M. R., & Chon, K.-S. (1989). Formulating and evaluating tourism policy using importance-performance analysis. Journal of Hospitality and Tourism Research, 13(3), 203–213. http://dx.doi.org/10.1177/109634808901300320 Fan, W. S., and Tsai, M. C. (2010). Factors driving website success: The key role of Internet customization and the influence of website design quality and Internet marketing strategy. Total Quality Management in Business, 21(11), 1141–1159. http://dx.doi.org/10.1108/17473611111114768 https://doi.org/10.1016/j.jairtraman.2014.11.005 https://doi.org/10.1016/j.procs.2019.11.233 https://doi.org/10.1016/j.jretai.2016.09.002 http://dx.doi.org/10.1016/j.asoc.2016.01.017 https://doi.org/10.30656/jsmi.v2i2.835 https://doi.org/10.1016/j.tourman.2004.06.008 http://dx.doi.org/10.1177/109634808901300320 Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 16 Ganesh, J., Reynolds, K.E., Luckett, M., & Pomirleanu, N. (2010). Online shopper motivations and e-store attribute: An examination of online patronage behavior and shopper typologies. Journal of Retailing, 86(1), https://doi.org/106-115. 10.1016/j.jretai.2010.01.003 Garibay, C., Pulido, H. G., & Montano, A. F. (2010). “Evaluation of a digital library by means of quality function deployment (QFD) and the Kano model. The Journal of Academic Librarianship, 36(2), 125–132. https://doi.org/10.1016/j.acalib.2010.01.002 Gregg, D. G., and Walczak, S. (2010). The relationship between website quality, trust and price premiums at online auctions. Electronic Commerce Research and Applications, 10(1), 1–25. https://doi.org/10.1007/s10660-010-9044-2 Goetsch, D. L., & Davis, S. B. (2014). Quality Management for Organizational Excellence: Introduction to Total Quality David L. Goetsch Stanley Davis Seventh Edition. Hausman, A.V., & Siekpe, J.S. (2009). The effect of web interface features on consumer online purchase intentions. Journal of Business Research, 62(1), 5-13. https://doi.org/10.1016/j.jbusres.2008.01.018 Ilbahar, E., Cebi, S. (2017). Classification of Design Parameters for E-commerce Websites: A Novel Fuzzy Kano Approach, Telematics and Informatics, DOI: https://doi.org/10.1016/j.tele.2017.09.004 Kim, S., & Jones, C. (2009). Online shopping and the moderating role of offline brand trust. Direct Marketing: An International Journal, 3(4), 282-300. https://doi.org/ 10.1108/17505930911000874 Kline, P. (1994). An Easy Guide to Factor Analysis, Routledge, London. Kuo, Y.F., Chen, J.Y., & Deng, W.J., 2012. IPA-Kano Model: a New Tool for Categorizing and Diagnosing Service Quality Attributes. Total Quality Management and Bussines Excellence, 23 (7-8), 731-748. https://doi.org/10.1080/14783363.2011.637811 Kotler, P., & Amstrong, G. (2014). Principles of Marketing (15th ed.). Essex: Pearson. Lee, H., S. Y. Choi, and Y-S. Kang. (2009). Formation of e-satisfaction and repurchase intention: moderating roles of computer self-efficacy and computer anxiety. Expert Systems with Applications: An International Journal 36 (4): 7848-7859. https://doi.org/ 10.1016/j.eswa.2008.11.005 Mansori, S., Cheng, Liat, B., & Hui Shan, L. (2012). A Study Of E-Shopping Intention In Malaysia: The Influence Of Generation X & Y. Australian Journal of Basic and Applied Sciences, 6(8), 28-35. Mona, A., Sepiden, F., Seyed, M.S.E. (2013). Linking trust, perceived website quality, privacy protection, gender, and online purchase intentions. IOSR Journal of Business and Management (IOSR-JBM). 13 (4), 63–72. Negash, S., Ryan, T., & Igbaria, M. (2003). Quality and effectiveness in Web-based customer support systems. Information and Management, 40(8), 757–768. https://doi.org/10.1016/S0378-7206(02)00101-5 Psaila, G., & Wagner, R. (2007). E-Commerce and Web Technologies: 8th International Conference, Regensburg, Germany, Proceedings. September 3-7. Rondovic, B., Cerovic, J., Durickovic, T. & Melovic, B. (2017). The Importance Of Observing The Difference In Website Evaluations Obtained From Different https://doi.org/106-115.%2010.1016/j.jretai.2010.01.003 https://doi.org/106-115.%2010.1016/j.jretai.2010.01.003 https://doi.org/10.1016/j.acalib.2010.01.002 https://doi.org/10.1007/s10660-010-9044-2 https://doi.org/10.1016/j.jbusres.2008.01.018 https://doi.org/10.1016/j.tele.2017.09.004 https://doi.org/10.1080/14783363.2011.637811 https://doi.org/10.1016/S0378-7206(02)00101-5 Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 17 Perspectives. International Journal for Quality Research, 11(2), 419-436. https://doi.org/ 10.18421/IJQR11.02-11 Sorensson, A., & von Friedrichs, Y. (2013). An importance-performance analysis of sustainable tourism: A comparison between international and national tourists. Journal of Destination Marketing & Management, 2(1), 14-21. Stuart, J. et. al (2003). Measuring Web site quality improvements: a case study of the forum on strategic management knowledge exchange. Industrial Management & Data Systems, 103(5), 297-309. Sugiyono. (2016). Metode Penelitian Kuantitatif, Kualitatif dan Kombinasi (Mixed Method). Bandung: Alfabeta. Vasić, N., Kilibarda, M., Andrejić, M., & Jović, S. (2020): Satisfaction is a function of users of logistics services in e-commerce, Technology Analysis & Strategic Management, https://doi.org/10.1080/09537325.2020.1849610 Wilson, N., & Christella, R. (2019). An Empirical Research of Factors Affecting Customer Satisfaction: A Case of the Indonesian E-Commerce Industry. DeReMa (Development Research of Management): Jurnal Manajemen, 14(1), 21. https://doi.org/10.19166/derema.v14i1.1108 Wilson, N., Keni, K., & Tan, P. H. P. (2019). The effect of website design quality and service quality on repurchase intention in the E-commerce industry: A cross- continental analysis. Gadjah Mada International Journal of Business, 21(2), 187–222. https://doi.org/10.22146/gamaijb.33665 Wilson, N. (2018). The impact of service quality and brand image toward customer loyalty in the Indonesian airlines industry. Jurnal Manajemen Indonesia 18 (3): 222- 234. Wilson, N., and R. Christella. (2019). Empirical research of factors affecting customer satisfaction: A case of the Indonesian e-commerce industry. DeReMa Jurnal Manajemen 14(1), 21- 44 Wilson, N., and K. Keni. (2018). Pengaruh website design quality dan kualitas jasa terhadap repurchase intention : Variabel trust sebagai variabel mediasi. Jurnal Manajemen dan Pemasaran Jasa, 11 (2): 291-310. Wong, M. S., Hideki, N., & George, P. (2011). The use of importance-performance analysis (IPA) in evaluating Japan’s e-government services. Journal of Theoretical and Applied Electronic Commerce Research, 6(2), 17–30. https://doi.org/10.4067/S0718-18762011000200003 Zhou, T., Lu, Y. B., and Wang, B. (2009). The relative importance of website design quality and service quality in determining consumers’ online repurchase behavior. Information Systems Management, 26(4), 327–337. © 2021 by the authors. Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1080/09537325.2020.1849610 https://doi.org/10.19166/derema.v14i1.1108 https://doi.org/10.22146/gamaijb.33665 Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 18 Appendix 1. Dimensions & Items Dimension Code Attributes Reference Website Design (WD) WD1 The website adequately meets my information needs Blut, M. (2016) WD2 The information on the website is effective Blut, M. (2016) WD3 The website is visually pleasing Blut, M. (2016) WD4 The display pages within the website are easy to read Blut, M. (2016) WD5 The text on the website is easy to read. Blut, M. (2016) WD6 The website loads quickly Blut, M. (2016) Product Quality (PQ) PQ1 This website has a good selection products Blut, M. (2016) PQ2 The site has a wide variety of products that interest me Blut, M. (2016) PQ3 All products on the website are available Blut, M. (2016) PQ4 The website offers discount product Blut, M. (2016) PQ5 The website has lower prices than offline stores Blut, M. (2016) Security Quality (SQ) SQ1 I feel safe in my transactions with the website Blut, M. (2016) SQ2 The website has adequate security features Blut, M. (2016) SQ3 This site protects information about my credit card Blut, M. (2016) SQ4 I trust the website to keep my personal information safe Blut, M. (2016) SQ5 It protects information about my web-shopping behavior Blut, M. (2016) Delivery Quality (DQ) DQ1 The product is delivered by the time promised by the seller Blut, M. (2016) DQ2 This website makes items available for delivery within a suitable time frame Blut, M. (2016) DQ3 It quickly delivers what I order Blut, M. (2016) DQ4 The website offers discount or free shipping Blut, M. (2016) DQ5 Seller provides delivery at low cost Vasic, et al. (2020) DQ6 Seller delivers products in accordance with the set conditions Vasic, et al. (2020) DQ7 The website offers the shipment tracking option Vasic, et al. (2020) Delivery Accuracy (DA) DA1 You get what you ordered from this website Blut, M. (2016) DA2 The website is truthful about its offerings Blut, M. (2016) DA3 The ordered products arrived in a good condition Blut, M. (2016) DA4 Delivers products in accordance with the set conditions Vasic, et al. (2020) DA5 Shipment content is seldom liable to complaints Vasic, et al. (2020) Customer Service (CS) CS1 This site provides a telephone number to reach the company Blut, M. (2016) CS2 This site has customer service representatives available online Blut, M. (2016) CS3 It offers the ability to speak to a live person if there is a problem Blut, M. (2016) CS4 It provides me with convenient options for returning items Blut, M. (2016) CS5 This site handles product returns well Blut, M. (2016) CS6 Return policy is simple Vasic, et al. (2020) Quality Improvement Of The E-Commerce Website Using Integration Of Kano Model-IPA With QFD Approach 19 Appendix 1. Dimensions & Items (Continue) Dimension Code Attributes Reference Customer Perception PS1 I am satisfied with this online retailer Blut, M. (2016) and Satisfaction (PS) PS2 The online retailer always meets my needs Blut, M. (2016) PS3 I consider this online retailer to be my first choice for next transactions Blut, M. (2016) PS4 I say positive things about this online retailer to other people Blut, M. (2016) PS5 I recommend this online retailer to someone who seeks my advice Blut, M. (2016) Appendix 2. How and What Matrix Shadrina & Ikatrinasari /Oper. Res. Eng. Sci. Theor. Appl. 4 (3) (2021) 1-20 20 Appendix 3. House Of Quality QUALITY IMPROVEMENT OF THE E-COMMERCE WEBSITE using integration of kano model-ipa with QFD approach Amalina Shadrina*, Zulfa Fitri Ikatrinasari 1. Introduction 2. Literature Review 2.1. Kano Model 2.2. Importance Performance Analysis (IPA) 2.3. Quality Function Deployment (QFD) 3. Methodology 4. Results 4.1. Validity and Reliability Analysis 4.2. Corelation Analysis 4.3. Hypothesis Analysis 4.4. Kano Model and Importance Performance Analysis 4.5. Quality Function Deployment 5. Discussions 6. Conclusion Reference