Companies invest significant resources in retaining their customers. Nonetheless, organizations have witnessed customer attrition due to inadequate loyalty. This trend is particularly prevalent among online customer bases. The root cause of this issue lies in the absence of an effective tool for measuring online customer satisfaction that surpasses the capabilities of existing methods. To address this concern, a quantitative study explored the dimensions of online customer satisfaction measurement and established a model applicable across industries for gauging and predicting online customer satisfaction. This was accomplished by conducting an online survey via SurveyMonkey with 384 respondents, employing supervised and unsupervised machine learning techniques in conjunction with the topic modeling algorithm, Latent Dirichlet Allocation (LDA). The findings of this study revealed a significant relationship between predictor variables such as navigation, playfulness, information quality, trust, personalization, and responsiveness and the target variable, online customer satisfaction, employing multiple linear modeling (LSM). Furthermore, it was observed that this phenomenon transcends age groups, impacting both younger and older customers alike. However, it is essential to acknowledge certain limitations, including the risk of overfitting, challenges in establishing external validity, a narrow focus on the retail sector (B2C), and a restricted scope limited to the United States market.