Classification of chemicals as skin sensitizers have traditionally relied on a small set of in vivo tests. Difficult to test substances, such as poorly soluble, mildly irritating, or those of Unknown or Variable Composition Complex reaction products or Biological Materials (UVCBs), producing weak or borderline results in Local Lymph Node Assay (LLNA), often benefit from additional data integration in a weight of evidence (WOE) approach. Advances in multiple testing and non-testing methodologies (in vivo, in vitro, and in silico) can now provide clarity and confidence in concluding on skin sensitization potential. Here we present several case studies using a WOE approach with difficult to test substances and highlight the utility of Toxicological Prioritization IndexTM (ToxPi™) as a comparative visualization and integration tool of toxicology studies. The three test chemicals chosen represent two poorly soluble substances, tetrakis (2-ethylbutyl) orthosilicate and decyl palmitate, and one UVCB substance, alkylated anisole. Data from in vivo and in vitro assays representing multiple key events within the skin sensitization adverse outcome pathway (e.g., direct peptide reactivity assay, human cell line activation test, GARD®Skin, LLNA) were either gathered from publicly available sources or specifically generated. Incorporating the data on our test chemicals as well as chemicals of a known sensitization class (sensitizer, irritating non-sensitizer, non-sensitizer) into ToxPi™ revealed biological activity profiles which were used to support class prediction for the three test chemicals. Using this method, the biological activity profiles for all three test chemicals were most consistent with a non-sensitizing class. This paper demonstrates that visualizing the WOE using mechanistic data maximizes the value of all data to the overall assessment of skin sensitization potential by reducing the uncertainty associated with any one individual assay.