How accurate are food waste tracking systems? Insights from healthcare and hotel kitchens

• Food waste tracking systems (FWTS): kitchen staff underreports food waste by ∼30 %. • Underreporting reached 80 % under lunch service in one healthcare kitchen. • Coverage and weighing frequency strongly correlate with reporting bias (R 2 > 0.80). • FWTS data appear precise but can be inaccurate due to enumerator effects (bias). • One healthcare kitchen showed negligible bias and high accuracy and reproducibility. Food waste tracking systems (FWTS) have become increasingly common as monitoring tools in the food service sector. Yet, staff-reported FWTS data are subject to uncertainties from enumerator bias, and their accuracy has rarely been investigated or empirically tested. This study provides insights into the reliability of FWTS data from three healthcare kitchens and one hotel kitchen. Using a staggered experimental design, staff-reported data (≥ 21 days) were compared with scientific control data (6 days) collected with the same FWTS under constant conditions. Staff-reported FWTS data underestimated food waste quantities by up to 80 %, with an average underreporting of approx. 29.4 % across mealtimes (breakfast, lunch, dinner) in the healthcare kitchens and approx. 30.7 % during breakfast buffet in the hotel kitchen. Our results also show that staff-reported FWTS quantities can shift the true mean values toward a biased underestimation without widening the confidence interval, making comparisons appear precise but inaccurate.