Abstract Background Clinical skill training (CST) is indispensable for first-year surgical residents. It can usually be carried out through video-based flipped learning (FL) within a web-based learning environment. However, we found that residents lack the process of reflection, blindly imitating results in losing interest and passion for learning in the traditional teaching pattern. The teaching method of "spot the difference" (SDTM), which is based on the fundamentals of the popular game of "spot the difference," is designed to improve students' participation and reflective learning during skill training. This study aimed to evaluate this novel educational model's short-term and long-term effectiveness for surgical residents in China. Methods First-year residents who required a three-month rotation in the head and neck surgery department were recruited to participate in a series of CSTs. They were randomized into SDTM and traditional FL (control) groups. Clinical skill performance was assessed with validated clinical skill scoring criteria. Evaluations were conducted by comparing the scores that contain departmental rotation skill examinations and the first China medical licensing examination (CMLE) performance on practical skills. In addition, two-way subjective evaluations were also implemented as a reference for the training results. Training effects were assessed using t tests, Mann–Whitney–Wilcoxon tests, chi-square tests, and Cohen’ s effect size (d). The Cohen’ s d value was considered to be small (<0.2), medium (0.2-0.8), or large (>0.8). Results The SDTM group was significantly superior to the control group in terms of after-department skill examination (t=2.179, p <0.05, d=0.5), taking medical history (t=2.665, p <0.05, d=0.59), and CMLE performance on practical skill (t=2.103, p <0.05, d=0.47). The SDTM members rated the curriculum more highly than the control on the items relating to interestingness and participation ( p < 0.05) with large effect sizes (d >0.8). There were no significant differences between the two groups on clinical competence (t=0.819, p =0.415, d=0.18), the first-time pass rate for CMLE (χ2 =1.663, p =0.197, d=0.29), and short-term operational skills improvement (t=1.747, p =0.084, d=0.39). Conclusions SDTM may be an effective method for enhancing residents' clinical skills, and the effect is significant both short- and long-term. The improvement effect seemed to be more significant in the peer-involved SDTM than training alone. However, despite positive objective results, SDTM still risks student learning burnout. Trial registration ISRCTN registry, ISRCTN10598469 , 02/04/2022，retrospectively registered.