X-ray powder diffraction (XRPD) is found consistently to be the most accurate analytical technique for quantitative analysis of clay-bearing mixtures based on results from round-robin competitions such as the Reynolds Cup (RC). A range of computationally intensive approaches can be used to quantify phase concentrations from XRPD data, of which the ‘full-pattern summation of prior measured standards’ (FPS) has proven accurate and parsimonious. Despite its proven utility, the approach often requires time-consuming selection of appropriate pure reference patterns to use for a given sample. As such, applying FPS to large and mineralogically diverse datasets is challenging. In the present work, the accuracy of an automated FPS algorithm implemented within the powdR package for the R Language and Environment for Statistical Computing was tested on a set of 27 samples from nine RC contests. The samples represent challenging and diverse clay-bearing mixtures with known concentrations, with the added advantage of allowing the accuracy of the algorithm to be compared with results submitted to previous contests. When supplied with a library of 201 reference patterns representing a comprehensive range of phases that may be encountered in natural clay-bearing mixtures, the algorithm selected appropriate phases and achieved a mean absolute bias of 0.57% for non-clay minerals (n = 275), 2.37% for clay minerals (n = 120), and 4.43% for amorphous phases (n = 14). This accuracy would be sufficient for top-3 placings in all nine RC contests held to date (RC1 = 2nd, RC2 = 2nd, RC3 = 1st; RC4 = 2nd; RC5 = 1st; RC6 = 3rd; RC7 = 3rd; RC8 = 1st; RC9 = 2nd). The comparatively low values of absolute bias in combination with the competitive placings in all RC contests tested is particularly promising for the future of automated quantitative phase analyses by XRPD.