In this Letter, via idealized and realistic case studies, we have documented the limitations of several conventional metrics in validating Cn2 profile predictions. We have introduced the (normalized) Kantorovich metric as a viable alternative and demonstrated its prowess.