Reliable rock classification is the key to identify target zones for successful hydraulic fracturing stimulation treatment in unconventional reservoirs such as organic-rich mudrocks. Such a rock classification scheme should take into account geologic attributes, petrophysical, and geomechanical properties (i.e., in situ stress gradient and elastic properties) to improve the likelihood of successful fracture treatment. However, conventional rock classification methods do not take into account stress gradients in the formation. We have developed a new rock classification technique that integrates four rock classification schemes based on the (1) geologic facies, (2) reservoir quality, (3) stress profile, and (4) completion quality. The techniques applied in these classification schemes include core description and thin section analysis, well-log-based depth-by-depth petrophysical and compositional characterization, and analysis of geomechanical measurements. Geomechanical analysis of core measurements and well logs provide a depth-by-depth assessment of minimum horizontal stress assuming vertical transverse isotropy in the formation. We have performed the geologic facies and reservoir quality classifications using an artificial neural network analysis, in which well logs and well-log-based estimates of the petrophysical and compositional properties were inputs to the network. Our technique was applied to a well located in the Wolfcamp Shale in the Delaware Basin. Based on the integrated rock classification results, we recommend the middle of the upper Wolfcamp and the bottom of the lower Wolfcamp depth intervals as the best candidates for fracture initiation and fracture containment zones, respectively. The selection of these zones was based on the reservoir quality and average minimum horizontal stress gradient calculated in these intervals. Our integrated rock classification technique can improve the planning and execution of completions design for hydraulic fracture treatments.