The paper addresses the need to consider anthropometric diversity in design and suggests a basic approach for the simultaneous consideration of variance in two key dimensions. This as a basic step from the common, but in many cases poor, approach to use univariate percentile data in design. The bivariate method described can be applied when utilising DHM tools for design in that key dimension values for extreme but likely anthropometric measurement combinations are calculated and entered as input data when representative manikins are defined. The mathematical procedure is described and the outcome of the method is compared to a typical percentile based approach, indicating more accurate accommodation levels being reached by the proposed method. The method is to be seen as a simple method to be used for basic design problems where variance in few anthropometric dimensions are to be considered simultaneously, and not as an alternative for more advanced multivariate methods. The paper takes a pragmatic standpoint, directing its message towards practitioners using DHM tools for design purposes.