This study consists of advanced text mining and Natural Language Processing (NLP) technique to analyse user reviews on commercial diet tracking apps, focusing on enhancing user engagement and satisfaction through persuasive System Design (PSD) model. By systematically categorising user feedback into areas such as primary task, dialogue, social and credibility, the research identifies key patterns and factors impacting user interactions, as well as provide deeper insight into how app features influence sustained user engagement and adherence to health goals. The categorised user feedback provides a distinct user reviews into four categorises which allow developers to pin point specific areas of where users are satisfied or requires specific refinements according to the four PSD models. The findings illustrate the diverse influences of PSD elements on user satisfaction and engagement. This methodological approach not only addresses a significant gap in understanding user feedback but also serves as pioneer attempt of using NLP incorporated with PSD model to refine health app features, thereby improving user outcomes and retention in specific areas of persuasive design models.