The objective was to evaluate the interest of yeast probiotic (Actisaf, Saccharomyces cerevisiae Sc47-CNCM I-4407, 10¹⁰cfu/g DM, Phileo Lesaffre Animal Care, France, 5 g/h/d) and yeast fraction (Safmannan, Phileo Lesaffre Animal Care, France, 10 g/h/d) distributed during 1 year to animals from 10 French commercial dairy farms, with 102 head/farm, on average. Salmonella has been identified in bulk milk at least 3 times over the previous year in the 20 selected farms (shedding results higher than 50,000 cfu/g manure) assigned randomly to 2 groups (C, YP) according to Salmonella spp. shedding before (visit 0) starting the project: C as control group; YP receiving yeast products daily. Farms were regularly monitored (visits 1 to 4) and analyzed on 1 composite environmental fecal sample (CF) and 6 individual fecal samples (IF) collected on randomly selected cows. Shedding level was evaluated using the most probable number or NPP count method and expressed in cfu/g. Results at visits 1 to 4 were compared with visit 0 using Mann-Whitney test. Multivariable analyses were conducted, using ordinary logistic regression (OLR) model and generalized estimating equation (GEE, family = binomial) model, to account for within-herd correlation. The median CF count was higher in YP (938,200) than C group (125,800) at visit 0. Difference between groups (YP – C) was less at both visits 1 and 4 (+155,000 and +104,343) and was negative at both visits 2 and 3 (−490 and −19,000). However, this trend remained insignificant using OLR analysis. The IF was higher in YP at visit 0 (median 53,000 versus 8,700; i.e., +44,300). Difference was smaller at visits 2, 3, 4 (+3,840, +4,000, +160) and was reversed at visit 1 (−9,800). A significant decrease in shedding was observed at visit 1, compared with visit 0, only in YC (58 cattle, P = 0.01). The GEE analysis was conducted on 122 animals presenting at least 3 counts over the 4 control points. The model showed an overall reduction of shedding in YP cattle, but the trend remained insignificant (coeff. −0.448, SD 0.674; adjusted odds ratio 0.64, 95% [0.17–2.39]). Consequently, feeding yeast products could help to reduce excretion of Salmonella on dairy farms. Further investigations are needed to confirm the results.