Significant fracture conductivity can be achieved using a much lower material cost based on the optimal partial-monolayer proppant concentration (OPPC) theory. However, experimental validation and investigation of the OPPC theory have been extremely rare in the literature. In this study, we used a laboratory fracture conductivity cell to conduct well-controlled fracture conductivity experiments to comprehensively study the role of effective stress, proppant size, rock type, and water soaking on the evolution of fracture conductivity as a function of increasing proppant concentration. With seven proppant concentrations (up to 2 lb/ft2) and seven effective stresses (up to 6,000 psi) used in the conductivity measurements, we experimentally confirmed that the correlation between fracture conductivity and proppant concentration was non-monotonic because of a competing process between fracture permeability and fracture width. We also investigated the influence of the above-mentioned experimental conditions on the OPPC and the corresponding optimal fracture conductivity (OFC). This is the first study that uses well-controlled laboratory experiments to comprehensively investigate non-monotonic fracture conductivity evolutions. The existence of the OPPC indicates that a relatively low proppant amount can be used to form a partial-monolayer proppant pack in the fracture space, which has similar or higher fracture conductivity compared to a multilayer proppant structure. This finding has important economic implications because high-strength, ultralight-weight proppant particles can be used to form partial-monolayer proppant packs in fractures, leading to sufficiently high fracture conductivity using a much lower material cost compared to multilayer proppant structures. Our experiments illustrated that proppant embedment is the primary mechanism that causes the competing process between fracture width and fracture permeability and consequently the non-monotonic fracture conductivity evolution as a function of increasing proppant concentration. Without proppant embedment, there will not be such a competing process, and the non-monotonic fracture conductivity evolution will not be observed.