The purpose of this study is to determine the significant effect of the application of word puzzle media on the beginning reading ability of students in class I of MIN 12 Medan. This study uses a quantitative approach to the experimental method with a Quasi-Experimental type. The quantitative research method using experiments has the aim of examining the effect of a particular treatment on the symptoms of a particular group compared to other groups.The design used in this study is the experimental and control class groups. The population and sample in this study are class IA students as the experimental class totaling 28 students using word puzzle media and class IB as the control class totaling 28 students using conventional learning media. The research instruments are tests and non-tests. In the pre-test in the experimental class, the mean (average value) was 58.39 while the post-test got a mean (average value) of 77.96 which was given learning treatment using puzzle media. Then in the control class, the pre-test mean (average value) was 52.61 and the post-test mean (average value) was 70.29 using conventional learning media.In the research data, it was analyzed using normality tests and homogeneity tests to determine whether the sample data was normally and homogeneously distri buted, then a hypothesis test (t-test) was conducted to determine the influence of word puzzle media on the initial reading abilities of students in grade I using the SPSS version 26 application. The hypothesis test conducted on the experimental and control classes yielded a sig. (2-tailed) value of 0.037 ≤ 0.05, thus rejecting Ho and accepting Ha. It can be concluded that the use of word puzzle media is more effective in teaching students to read and has an influence.