A computational strategy is presented that allows rapid implementation of genetic evaluations using multivariate mixed models. Data generated in different testing programs such as field tests of boars and gilts, litter recording schemes and station tests of sibs may be combined to provide an estimate of the aggregate genotype. Residual and additive genetic covariance structures are given for the multivariate evaluation of individual measurements and group averages because they often are collected for sib groups at test stations using a modified animal model. Pseudo code is given for the implementation of a "generic" testing structure illustrated by a numerical example based on six traits from field tests of boars and four traits from station tests of sibs. BLUPs for all six traits are calculated for boars, parents and sib groups. Aggregate genotypes that correspond to the selection indices commonly used are calculated for selection candidates.