Designing the mechanism of online businesses to prevent customers post purchase regret

EXTENDED ABSTRACT
Introduction: The current research was conducted with the aim of designing the mechanism of online businesses to prevent post purchase regret behavior of customers using a mixed exploratory method. Regret is a feeling that occurs in the purchase process or after the purchase, and this feeling can be caused by not meeting the initial expectations of the customer. The results of a research conducted in the UK show that about 82% of adults have experienced shopping aversion in the past. Also, according to a survey conducted by Slickdeals in 2022; 74% of Americans have regretted after online shopping. Regret behavior after online shopping is sometimes different and more complicated than offline shopping behavior. Post purchase Regret can be due to individual characteristics; personality traits; product compatibility with customer expectations; customer shopping experience; Confidence in online shopping; time pressure on the customer to buy; Perceived risk in online shopping or even lack of sleep. In addition, when the consumer feels that the product of competing companies is better than the purchased product in terms of price and features or when there is a difference between the previous evaluation of the product and the final purchase, a feeling of regret is created. According to what has been described, so far, various studies have been conducted to investigate the regret behavior of customers after online shopping, but research that comprehensively and systematically considers this behavior has been less investigated. Therefore, the effort of the present research is to design the mechanism of online businesses with a mixed exploratory approach to prevent post purchase regret in online shopping among Iranian customers.
Methodology: In the qualitative part, the research strategy of Grounded theory and the systematic design of Strauss and Corbin have been used for theorizing. In this plan, the data obtained from the interviews are placed in pre-determined sections, i.e. causal conditions, core category, context, intervening conditions, strategies and consequences. The data collection tool was a semi-structured interview and its questions were designed based on the components of the paradigm model. Data triangulation strategy was used to conduct the interviews and 14 experts including university professors, online business industry activists and web businesses were interviewed through judgmental sampling. In order to check validity, research members who included 4 university professors and a Ph.D. student reviewed the coding process and their points of view were applied in the coding process. Retest method was used to calculate reliability. The reliability method of two evaluators was used to calculate the reliability. In this way, among the conducted interviews, 3 interviews were selected as samples and evaluated by another researcher. Cohen's kappa coefficient was used to calculate reliability. Kappa coefficient equal to 0.811 was obtained, which indicates almost complete agreement between two evaluators. In the quantitative part, structural equation modeling was used to evaluate and validate the designed model, and SmartPLS software was used to perform its calculations. The statistical population was the customers of the Digikala online store who made online purchases from the home appliances department from 2013 onwards and registered their complaints in the company's complaints department. Using simple random sampling, 800 emails were sent to customers and 396 analyzable questionnaires were obtained. For the validity of the quantitative part of the research, confirmatory factor analysis, convergent and divergent validity were used. For reliability, composite reliability and Cronbach's alpha were used.
Discussion and Results: In conducting the research, after implementing the interviews, their key points were determined and the primary codes of each were extracted. The interviews continued until the data reached theoretical saturation and no new points were observed in the interviews. After conducting 14 interviews, 457 codes were obtained as primary codes, which created 71 separate concepts by removing similar codes. In the axial coding stage, one of the categories obtained from open coding was identified as the category that is the basis of the research theory. Then other categories, i.e. causal conditions, context and intervening conditions, strategies and consequences were related to it. In this way, 71 obtained concepts were classified and placed in 20 categories. In the selective coding stage, the 20 extracted categories were linked to each other and the paradigm model of the research was designed using MAXQDA software.
In the quantitative part, structural equations used to validate the model. First, in order to check the normality of the data, the Kolmogorov Smirnov test was used. The results showed that the data is not normal, so the PLS method was used to estimate and test the model. To evaluate the confirmatory factor analysis model, there are several fit indices that must be interpreted together. In this research, Goodness of Fit index, Normed Fit Index and Standardized Root Mean Square Residual were used to evaluate the confirmatory factor analysis model. In this research, the GOF is equal to 0.567, the NFI is equal to 0.848, and the SRMR index equal to 0.061. In general, according to all three fit indices, it can be said that the data of this research has a good fit and the questions are aligned with the theoretical framework.
Conclusion: The model obtained from the present research provides a comprehensive view to the companies active in the field of online business so that they can make decisions regarding the appropriate marketing and sales strategies by knowing the buying behavior of online customers. Researchers who intend to study the behavior of post purchase regret in the online environment can study regret in purchases goods that are likely to have more regret. For example, possible regret in choosing luxury and expensive goods, or cheap goods that have low quality. Also, the evaluation of post purchase regret in products that affect the environment. Investigating the effect of discounts offered on sites and putting time pressure on customer are other areas that are recommended to researchers.