The purpose of this study is to investigate how genres vary and figure out the factors that generate genre variation. The quantitative multi-dimensional analysis is used to examine genre variation of cross-border e-commerce English in South Asia. The texts in the observed corpus collected from four country websites of Daraz, a significant cross-border e-commerce platform, were tagged and analyzed using Multidimensional Analysis Tagger and SPSS statistical software. The results of linear regression analysis, independent sample t-test and Analysis of Variance show salient differences between the observed corpus and reference corpus. The research also indicates that four sub-corpora from country Websites are brought into line with each other. They show salient differences (p<0.05) with the general corpus in two dimensions of linguistic variation. The findings indicate that variables that lead to the cross-border e-commerce English genre variation might be attributable to cultural backgrounds, regional market sizes, and fundamental internet facilities. In conclusion, these findings lend significant empirical support to systemic functional theories and suggest further research in the application of a multi-dimensional analysis approach in the cross-border e-commerce English genre.