Thailand is transitioning into a super-aged society, and its growing elderly population faces an increased risk of postoperative cognitive dysfunction (POCD). Despite its association with reduced quality of life and heightened mortality, POCD remains underdiagnosed. This study aimed to evaluate the incidence of POCD using a web-based Montreal Cognitive Assessment application (Auto-MoCA) at 1 week postoperatively. We also compared characteristics between patients with and without POCD and examined recovery patterns at 3 months. We conducted a prospective, descriptive cohort study at a university hospital in Bangkok, enrolling Thai-speaking patients aged ≥ 60 years who underwent major elective surgery. Cognitive function was repeatedly assessed via the Auto-MoCA application at baseline, 1 week, and 3 months. POCD was defined as a decrease of ≥ 2 points from baseline. Among 169 enrolled patients, 111 completed the 1-week assessment. POCD occurred in 40.5%, with significant declines in visuospatial, attention, language, abstraction, and orientation domains. Cardiovascular disease was a significant risk factor. By 3 months, 85 patients repeated the Auto-MoCA test. Of these, 34 (40.0%) had developed POCD compared to their baseline results, with 42.9% of affected patients recovered, primarily demonstrating improvements in visuospatial, attention, and delayed recall domains. Postoperative cognitive dysfunction (POCD) was common in older adults, affecting over 40% of patients at both 1 week and 3 months after surgery. With Cardiovascular disease emerged as a significant risk factor. However, nearly half of those affected showed cognitive recovery by 3 months. Underscoring the utility of digital cognitive assessment tools, such as the Auto-MoCA application, and highlight the need for future interventional research aiming to facilitate cognitive recovery in older adults following surgery.