This paper firstly proposes a precise vocal music teaching model based on information technology. Through the collection and in-depth mining of educational big data such as class attendance, after-class singing training, students’ performance, etc., and the use of precise mathematical model for processing and analysis, students can all explore deeply according to their own learning rhythm and interest points, and enhance the effect of vocal music learning. Then, precision teaching is applied to Z University, using a questionnaire survey and comparative analysis to discuss its application effect. Finally, the optimization strategy for precision teaching reform in the age of information technology is presented. The study shows that there is a significant linear relationship between class attendance, after-class training, and performance (P=0.001). The precise teaching method can significantly promote students’ development and improve their performance. There is a significant difference between the performance of the experimental and control classes with Sig. value equal to 0.03. Most students think that precision teaching is better than traditional teaching and that it is becoming a big trend.