Accurate quantification of farm-level actual evapotranspiration (ETa) remains one of the major data constraints in irrigation water management of South African irrigated environments. This study evaluates the performance and application of a smart field weighing lysimeter (SFL-600) which is the first of its kind in the country. The lysimeter was used for field-scale monitoring of ETa and irrigation efficiency across four cropping seasons involving barley, maize and soybean. Continuous measurements of lysimeter mass, drainage and soil-water parameters were measured at 1-min temporal resolution. Complementary meteorological data was measured from an automatic weather station to compute reference evapotranspiration (ETo). The actual evapotranspiration was derived using a water-balance approach. The measured data was filtered using Savitzky–Golay smoothing technique to remove mechanical noise. The quantified ETa was validated against ETa estimated from a soil-moisture-balance method. Results demonstrated a strong correlation between the two ETa methods with correlation values: r = 0.91 and 0.96 which confirmed the reliability of the smart lysimeter for direct ETa assessment. Irrigation deficit and surplus analysis revealed substantial mismatches between water applied and crop water use, this demonstrated an inefficient irrigation scheduling in the study area. Locally derived crop coefficients (Kc) differed slightly from the FAO-56 values, demonstrating the importance of site-specific calibration for improving ETc estimation. The integrated sensor system provided additional insights into the crop root-zone dynamics which complemented ETa measurements. The findings of this study demonstrate the capability of smart weighing lysimeters to generate accurate ETa data for improving irrigation scheduling and for supporting the calibration of indirect approaches and ET models in water scarce and regions with limited observational networks.