Advances in acquisition and processing technology help overcome imaging challenges in complex structural settings. The widespread adoption of wide-azimuth (WAZ) and the move towards full-azimuth (FAZ) acquisition geometries, both combined with increasing offsets, result in significantly improved illumination. Reduced compute cost and improved performance enabled reverse time migration (RTM) to emerge as the imaging algorithm of choice in such settings. Of course, an accurate velocity model is a key component in realizing the full potential of these acquisition geometries and algorithms. The trend is towards increasingly more complex anisotropic models, with a move from vertical transverse isotropy (VTI) to tilted transverse isotropy (TTI) and even orthorhombic. In the Gulf of Mexico (GoM), though the importance of defining an accurate anisotropic model in the supra-salt section cannot be understated, the largest contributing factor to a good image subsalt is often the correct delineation of the “salt body” itself. Without an accurate definition of the salt geometry, the subsalt image invariably remains distorted and poorly resolved. In this paper, we will focus on this portion of the depth imaging workflow and illustrate how the techniques of RTM scenario testing and image partitioning can be used in combination to both help define the salt geometry and improve the final post-migration image. We will describe a practical workflow and the key components that we feel are necessary for its success. In addition, we will illustrate a number of lessons learned during the course of recent projects executed in the GoM.