The invention relates to the technical field of big data, in particular to a site selection method, and aims at solving the problem that existing site selection methods are long in time consumption, high in cost, inaccurate and difficult to understand, and do not consider mutual exclusion factors. The site selection method specifically comprises the following steps of: 1, determining store type which needs site selection; 2, dividing grids; 3, analyzing grid elements; 4, determining a site selection equation; 5, verifying accurate of the site selection equation; and 6, completing site selection according to the site selection equation. The method is suitable for the store site selection in the fields of commercial retail industry, logistics industry, conference and exhibition industry, financial industry, catering industry, tourist industry, real estate industry, advertising industry, information consulting service industry, public utilities and the like.