Whole-blood donors are at increased risk for iron deficiency and anaemia. The current standard of haemoglobin monitoring is insufficient to ensure the maintenance of proper iron reserves and donor health. We aimed to determine the effects of ferritin-guided donation intervals for blood donor health and blood supply in the Netherlands. In this stepped-wedge cluster-randomised trial (FIND'EM), the 138 fixed and mobile donation centres in the Netherlands are organised into 29 geographical clusters and the clusters were randomly assigned to four treatment groups, with two groups being further split into two per a protocol amendment. Eligible donors were whole-blood donors who consented for use of their leftover material in the study. Each group was sequentially crossed over from the existing policy (haemoglobin-based screening; control) to a ferritin-guided donation interval policy over a 3-year period. In the intervention groups, in addition to the existing haemoglobin screening, ferritin was measured in all new donors and at every fifth donation in repeat donors. Subsequent donation intervals were extended to 6 months if ferritin concentrations were 15-30 ng/mL and to 12 months if they were less than 15 ng/mL. Outcomes were measured cross-sectionally across all donation centres at four timepoints. Primary outcomes were ferritin and haemoglobin concentrations, iron deficiency, and haemoglobin-based deferrals. We assessed all outcomes by sex and menopausal status and significance for primary outcomes was indicated by a p value of less than 0·0125. This trial is registered in the Dutch trial registry, NTR6738, and is complete. Between Sept 11, 2017, and Nov 27, 2020, 412 888 whole-blood donors visited a donation centre, and we did measurements on samples from 37 621 donations from 36 099 donors. Over 38 months, ferritin-guided donation intervals increased mean ferritin concentrations (by 0·18 log10 ng/mL [95% CI 0·15-0·22; p<0·0001] in male donors, 0·10 log10 ng/mL [0·06-0·15; p<0·0001] in premenopausal female donors, and 0·17 log10 ng/mL [0·12-0·21; p<0·0001] in postmenopausal female donors) and mean haemoglobin concentrations (by 0·30 g/dL [95% CI 0·22-0·38; p<0·0001] in male donors, 0·12 g/dL [0·03-0·20; p<0·0074] in premenopausal female donors, and 0·16 g/dL [0·05-0·27; p<0·0044] in postmenopausal female donors). Iron deficiency decreased by 36-38 months (odds ratio [OR] 0·24 [95% CI 0·18-0·31; p<0·0001] for male donors, 0·49 [0·37-0·64; p<0·0001] for premenopausal female donors, and 0·24 [0·15-0·37; p<0·0001] for postmenopausal female donors). At 36-38 months, haemoglobin-based deferral decreased significantly in male donors (OR at 36-38 months 0·21 [95% CI 0·10-0·40, p<0·0001]) but not significantly in premenopausal or postmenopausal female donors (0·81 [0·54-1·20; p=0·29] and 0·50 [95% CI 0·25-0·98; p=0·051], respectively). Ferritin-guided donation intervals significantly improved haemoglobin and ferritin concentrations and significantly decreased iron deficiency over the study period. Haemoglobin-based deferrals decreased significantly for male donors, but not female donors. Although this intervention is overall beneficial for maintenance of iron and haemoglobin concentrations in donors, increased efforts are needed to recruit and retain donors. The Sanquin Research Programming Committee.