APP provides users with work and entertainment convenience, but as it diversifies, people have more choices and pay more attention to the product’s intuitive feel. This paper proposes an improvement method for the KANO model while projecting the demand item weights of APP users. Text mining and sentiment analysis algorithms are utilized to obtain the emotional psychology of users’ experiences when using APP services. Combined with the psychological characteristics of the users, based on the current situation of the user group, we construct a user research framework to determine the user group of the APP service designed in this paper and establish the corresponding user portrait. The design strategy of an APP service is proposed, and a feasibility assessment system is designed to analyze user satisfaction with the APP using experience. The sensitivity r-value is 0.73245, and there are 10 elements that need to be improved, with charm-type elements accounting for 4 and expectation-type elements accounting for 6, respectively. According to the optimization strategy, the APP service design is optimized and improved, and the performance of the improved APP service is empirically examined; the satisfaction before and after filling in the missing values is 3.645 and 3.561, respectively, which is in the range of “general” and “comparative” satisfaction. The feasibility of the optimization scheme is evident.