Dashboard for Location Segmentation in New Student Admission Promotion

The intense competition for new student admissions among universities demands a strategic approach to attract applicants effectively. Telkom University Surabaya has traditionally promoted itself by sending employees and student representatives to various schools. However, the admissions team has not yet segmented school locations to prioritize those with the highest potential. This study addresses this gap by clustering schools based on their potential contribution to new student admissions using K-Means Clustering with Principal Component Analysis (PCA). The dataset consists of new student admissions data from 2020 to 2024, specifically focusing on students who purchased registration PIN. Variables include school category, district, province, registration path, PIN price, school type, domicile city, study program, school island, and faculty. The K-Means clustering analysis, after preprocessing, identified the optimal number of clusters as 3, confirmed through the Elbow Method, Silhouette Score, and Davies-Bouldin Score. The three clusters illustrate different levels of student interest, allowing the admissions team to target schools across various cities, primarily focusing on East Java, where the highest applicant numbers are concentrated in areas like Surabaya, Sidoarjo, and Malang. Visualization through three-dimensional PCA components provided clear distinctions among clusters, with one cluster showing particularly high variance in student interest, mainly for popular programs such as Informatics and Digital Business. This clustering outcome serves as a valuable tool for the university’s admissions team to develop a targeted and efficient promotional strategy. By identifying clusters with high potential, Telkom University Surabaya can allocate resources toward schools more likely to generate applicants. Additionally, strategic geographic areas are pinpointed for promotion, enabling the university to increase new student numbers in key regions for future admissions cycles.