The purpose of this study was to evaluate the EpiDerm™ Corrosivity (OECD 431) and Corrositex® Time Monitor (OECD 435) as in vitro methods to predict skin corrosivity for extreme pH (≤2 or ≥ 11.5) products. Extreme pH can be a useful predictor of irritation but may lead to over classification in weakly buffered systems. Hazard classification guidelines such as the Globally Harmonized System of Classification and Labeling of Chemicals (GHS) recommend testing with a validated in vitro method to confirm a non-corrosive classification for an extreme pH product. Our objective was to identify a method that could accurately identify corrosive and non-corrosive alkaline products. 10/15 products tested on the Epiderm assay predicted the same skin classification when compared with the in vivo data. The remaining five formulas over-predicted the skin classification when compared with the in vivo data. There were no products in which the Epiderm under-predicted the skin classification when compared to the in vivo results. The Corrositex assay was able to accurately predict 3/7 formulas, four products were over predicted as corrosive by Corrositex; none were under predicted. Epiderm results were also compared to classifications made under JDI’s internal hazard assessment process which bridges new formulas to classification guidelines based on an extensive database of historical in vivo data on specific chemical and formulation categories. This weight of evidence approach is consistent with GHS guidance on use of professional judgment to classify a product. JDI’s process accurately predicted the in vivo data 14/18 times. Two products were under predicted and two over predicted. The EpiDerm assay is a promising alternative to the use of animals to confirm non-corrosive classifications; however limitations were seen with higher solvent levels. The Corrositex assay did not reliably identify non-corrosive formulations.