This research aims to (1) develop a web application for attendance recording using facial recognition for teachers and staff at the kindergarten; (2) evaluate web application performance and (3) evaluate user satisfaction with the web application. The identified problem is that the school’s existing time recording system relies on a time clock machine, which often breaks down. The process of summarizing the records is time-consuming, and when switching to manual recording, errors frequently occur, resulting in delays in payroll processing. Moreover, teachers arriving late affect both the quality of student supervision and student safety. A total of 11 staff members were selected to assess the application's efficiency, and 40 users were selected to assess the application's satisfaction including all staff members from the kindergarten. The research tools included a user satisfaction questionnaire, and statistical analysis was conducted using mean and standard deviation.
            The results showed that (1) the web application accurately identified all staff members; (2) the web application had an accuracy of 88.02% and (3) overall user satisfaction was at a [level], with a mean score of 3.70 and a standard deviation of 1.72 from the users at the kindergarten and a mean score of 4.56 and a standard deviation of 0.64 from users at Computer Department, Chiang Mai Rajabhat University. These findings confirm that the web-based facial recognition attendance system is effective and can be practically used for attendance tracking.