The importance of saturated hydraulic conductivity (Ksat) as a soil hydraulic property led to the development of multiple pedotransfer functions for estimating it. One approach to estimating Ksat uses textural classes rather than specific textural fraction contents as a pedotransfer input. The objective of this work was to develop and evaluate a grouping-based pedotransfer procedure to estimate Ksat for sample sizes used in laboratory measurements. A search of publications and reports resulted in the collection of 1245 data sets with coupled data on Ksat, USDA textural class, and bulk density in the United States into a database called USKSAT. A separate database was assembled for the state of Florida that included 24,566 data sets. Data in each textural class were split into high and low bulk density groups using the splitting algorithm that created the most homogeneous groups. Sample diameters and lengths were <10 cm. Peaks of the semi-partial R2 were well defined for loamy soils. The threshold bulk density separating high and low bulk density groups is 1.24 g cm-3 for clay soils, about 1.33 g cm-3 for loamy soils, and about 1.65 g cm-3 for sandy soils. The high bulk density groups included a much broader range of Ksat values than the low bulk density groups for clays and loams but not sandy soils. Inspection of superimposed dependencies of Ksat on bulk density in the USKSAT database and in the Florida database showed their similarity. When geometric means were used as estimates of Ksat within groups, the accuracy was not high and yet was comparable with estimates obtained from far more detailed soil information using sophisticated machine learning methods. Estimating Ksat from textural class and bulk density may have the advantage of utility in data-poor environments and large-scale projects. © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA.