Satellite-based quantitative precipitation estimates (QPE) offer the potential for global, near real-time monitoring of precipitation. Provided their accuracy, in terms of frequency and intensity structures, can be verified, such products would prove to be highly valuable for constraining uncertainty in land data assimilation, hydrological simulation and short-term prediction applications. Two gauge-corrected and three uncorrected satellite-based QPE products are assessed over México against a new composite gauge dataset developed from data collected during the 2004 North American Monsoon season. Analysis of daily averaged rain rates, rain-rate conditional biases, and frequency maps each show a tendency for uncorrected satellite QPE products to overestimate the frequency of moderate to heavy precipitation events (>25 mm/d) with respect to gauge-only analyses. While all products reasonably captured the large-scale distribution of rainfall, some uncorrected products, particularly those emphasizing infra-red based retrieval of rain rates, possessed comparatively low pattern correlation scores with the gauge composite. Although gauge-corrected products tended to somewhat underestimate rainfall at heavy event thresholds, significant value, in terms of overall bias correction, appears to be added to gauge-corrected QPE products versus uncorrected products. This added value, however, highlights ongoing challenges with regards to collecting and integrating surface gauge data in an operational QPE framework.