The AIAA CFD Drag Prediction Workshop (DPW) Series was initiated in 2000 by a working group of members from the Applied Aerodynamics Technical Committee of the American Institute of Aeronautics and Astronautics. Since then, numerous workshops have been held providing participants a unique way to compare their results on several different test cases. The 7th Drag Prediction Workshop (DPW-7) focuses on predicting the effect of shock-induced separation on the variation of lift and pitching moment with increasing angle-of-attack at transonic conditions. The evaluation of this phenomenon on the NASA Common Research Model Wing-Body (CRM) configuration is rather challenging for solvers based on the Reynolds-Averaged Navier-Stokes (RANS) equations. In this presentation, it is proposed to show the benefits of anisotropic mesh adaptation, even in the presence of separated flows. Previous works highlighted that it is possible to achieve very accurate prediction on fully-unstructured adapted meshes composed only of tetrahedra. Moreover, the same accuracy can be obtained with a reduction of the mesh size by a factor between 20 and 100 with respect to the best-practice expert-created meshes. Numerous computations from the DPW-7 have been carried out and grid convergence was obtained thanks to anisotropic mesh adaptation on both steady and alpha sweep configurations. Goal-oriented mesh adaptation provided much better results than their feature-based counterparts and proves to be the proper strategy to tackle this kind of configuration.