Current agronomic practices for potato production in the irrigated areas of southern Alberta involve a hill/furrow configuration that was adopted from elsewhere, and designed to shed rainfall away from the hill and into the furrow. However, the principal intent of supplemental irrigation is to capture as much of the applied water into the hill, where the potato tubers and roots are located, and minimize water accumulating in the furrow. A three-year project began in 2006 to quantify the potential irrigation water savings of altered hill shapes for potato production. The three treatments (standard hill, flat-topped hill, and double-planted wide-bed) were arranged in a randomized strip plot design replicated four times. Soil water in each treatment was generally kept between 60 and 90% of available. A fourth treatment, triple-planted wide bed, was added to the project in 2008. The irrigation requirements to maintain the treatments were 487, 442, and 449mm for the standard hill, flat-topped hill, and double-planted bed, respectively, in 2006 and 442, 408 and 411mm for the same treatments in 2007. This translates into approximately 10% less irrigation water required for the flat-topped hill shape compared to the standard hill shape. The flat-topped hill shape required 5.0% more irrigation than the standard hill in 2008, but the double and triple-planted wide beds required 8.0 and 9.9%, respectively, less irrigation water than the standard. Although not always statistically significant, water use efficiency was greater in all years for the altered bed shapes compared to the standard hill geometry. Greater water use efficiency can be interpreted as more of the applied water infiltrated into the hill, where the potato plant could use it for transpiration and tuber development. Total yield was greater in 2006 for both the flat-topped hill (72.3Mgha⁻¹) and wide-bed hill (69.2Mgha⁻¹) compared to the standard hill (61.4Mgha⁻¹); however, the treatments were not significantly different. Significantly greater marketable yield was realized from the flat-topped hill treatment in 2006. This treatment also had a significantly greater number of marketable size tubers. In 2007, there were no significant differences in total yield; however, the standard and flat-topped treatments had a significantly greater number and yield of tubers in the 113-170g size category. Significant differences in total yield were found in 2008. The triple-planted wide bed had significantly greater yield in the smaller size categories compared to the standard treatment and significantly greater total tuber numbers than the other treatments, but the increase was in the smaller size categories, less than 170g. There were no significant differences among the treatments in yield or total number of tubers in the size categories greater than 171g in 2008.