Permanent mold casting produces the second most aluminum cast parts among all casting processes. In this process, mold coating changes the heat transferred from the molten metal to the mold by acting as an insulating layer. Moreover, the coating thickness is a significant variable regarding the coating’s thermal resistance, which strongly influences the microstructure of cast parts and the thermal shock on expensive molds. However, in casting production, coating peeling-off and repeated recoating result in an inhomogeneous coating thickness distribution. Due to the high working temperatures of the molds, no efficient online coating thickness measurements exist. We propose an indirect evaluation method based on the as-cast surface corresponding to the coating area. Our experiments analyzed as-cast and coating surfaces at nine different coating thicknesses. The results show a close correlation between the as-cast surface roughness parameter of arithmetical mean height Sa and the coating thickness. Based on this correlation, we can derive the coating thickness from the corresponding as-cast surface analysis. Furthermore, the coating peel-off area and other casting surface defects are easily recognized in these surfaces. In our next work plan, an affordable optical camera and proper light conditions will be tested by taking photos of as-cast surfaces, and an algorithm for the real-time automatic evaluation will be developed.