This paper presents a novel three-step procedure, called MICOM, to assess measurement invariance in composite models using variance-based structural equation modeling (SEM), such as partial least squares (PLS) path modeling. The procedure addresses the lack of established methods for assessing measurement invariance in composite models, which are dominant in variance-based SEM. The MICOM procedure involves three steps: configural invariance, compositional invariance, and equality of composite mean values and variances. The first step ensures that the same basic factor structure exists across groups. The second step checks whether the composite is formed equally across groups by comparing the correlation between composite scores using weights from different groups. The third step assesses whether the composite's mean values and variances are equal across groups. The procedure is validated through a simulation study and applied to an empirical example. The results show that the MICOM procedure appropriately identifies no, partial, and full measurement invariance. The study highlights the importance of establishing measurement invariance before conducting multigroup analyses in international marketing and other disciplines. The MICOM procedure provides a standard means to assess measurement invariance in composite models. Keywords: Methodology, Structural equation modelling, Measurement, Measurement invariance, Partial least squares, MICOM, Multigroup, Variance-based SEM, Composite models, Permutation test, Path modelling.This paper presents a novel three-step procedure, called MICOM, to assess measurement invariance in composite models using variance-based structural equation modeling (SEM), such as partial least squares (PLS) path modeling. The procedure addresses the lack of established methods for assessing measurement invariance in composite models, which are dominant in variance-based SEM. The MICOM procedure involves three steps: configural invariance, compositional invariance, and equality of composite mean values and variances. The first step ensures that the same basic factor structure exists across groups. The second step checks whether the composite is formed equally across groups by comparing the correlation between composite scores using weights from different groups. The third step assesses whether the composite's mean values and variances are equal across groups. The procedure is validated through a simulation study and applied to an empirical example. The results show that the MICOM procedure appropriately identifies no, partial, and full measurement invariance. The study highlights the importance of establishing measurement invariance before conducting multigroup analyses in international marketing and other disciplines. The MICOM procedure provides a standard means to assess measurement invariance in composite models. Keywords: Methodology, Structural equation modelling, Measurement, Measurement invariance, Partial least squares, MICOM, Multigroup, Variance-based SEM, Composite models, Permutation test, Path modelling.