Purpose: The purpose of this paper is to examine how a graduate institute at National Chiayi University (NCYU), by using a model that integrates analytic hierarchy process, cluster analysis and correspondence analysis, can develop effective marketing strategies. Design/methodology/approach: This is primarily a quantitative study aimed at developing a marketing mix for a graduate institute at NCYU in Taiwan. A survey using stratified random sampling was conducted, with 14 universities from four different areas in Taiwan randomly selected for the study. Two questionnaires were conducted: a Likert's five-scale questionnaire regarding school images and an analytic hierarchy process (AHP) questionnaire regarding school selection factors were administered to 640 undergraduate students. Of the total number of questionnaires, 602 (94 percent) valid school image questionnaires and 570 (89 percent) valid school selection factors questionnaires were used. Findings: The results of AHP revealed that the five most important factors for students' school selection were: employability, curriculum, academic reputation, faculty, and research environment. The results of clustering analysis identified five student groups for market segmentation, and they are the Prominence group, the Less aware group, the Pragmatic group, the Austerity group, and the Fastidious group. Finally, the results of correspondence analysis suggested that students of the Pragmatic Group are more likely to be attracted by NCYU, and also, students perceived NCYU to be strongly associated with lower tuition, fewer entrance-exam subjects, lower entrance-exam pass rates, and easier graduation requirements. Research limitations/implications: It would be better to conduct a factor analysis before using AHP. Practical implications: Particularly, NCYU should establish new curricula relevant to internationalization, develop curricula in school finance and educational economics, and form study groups to enhance graduating student employment opportunities. Generally, higher educational institutions may adopt the research model developed in this study to develop their marketing mix for better results. Originality/value: This paper documents research that was the first to integrate AHP, cluster analysis, and correspondence analysis in developing a marketing mix for higher educational institutions. (Contains 2 tables and 6 figures.)