The purpose of this paper is to establish the modified IPA (Importance-Performance Analysis) methodology for service quality characteristics and improvements in the hotel industry. Traditional IPA methodology has its shortcoming as it assumes that importance-performance is a normal distribution. However, if that is not the case, the traditional IPA model cannot correctly identify the priority for improvements. Therefore, this paper runs Box-Cox Transformation to convert data into normal distributions, in order to perform statistical analyses and comparisons on a like- for-like basis. This approach avoids the potential problem of the IPA model but maintains its simple and intuitive decision-making model. This paper examines standard hotels and performs literature reviews and experts’ interviews in order to establish the list of service quality characters for the hotel industry. It also conducts a survey to illustrate the application and benefits of the modified IPA method, so as to identify the key characteristics of the service quality and the ways to improve service quality of standard hotels.