Undergraduate Candidate Experience and Engagement: Insights from a Case Using CRISP-DM and Machine Learning

Several channels throughout their journey choose a course to enroll. The institutional website is among these channels. The way it is designed might influence how engaged these visitors are. Web analytics tools allow collecting high amounts of user behavior data, which can generate insights that help to improve HEI website and then incentivizing prospective students to apply for a course. Techniques of data mining, with CRISP-DM method, were used to help generating insights with an applied HEI case study. The expected outcome of this research was to unveil the most relevant segment of users and behaviors. The tools applied to collect, store, transform, visualize and clustering model data using X-Means algorithm were Google Tag Manager, Analytics, BigQuery, Data Studio, and RapidMiner. The main results showed four relevant groups of users. The key outtakes were that course and course unit pages were relevant in terms of attracting volume of users, but not in inciting engagement. In comparison, the homepage and the “general undergraduate course page” brought less users and their sessions usually lead to more engaged experiences. This study presented insights of the impact of brand awareness and landing pages design on engagement rates of the HEI website.