Abstract In bakeries, a number of operations such as mixing are associated with exposure to air-suspended flour dust and related bioburden. The aim of this study was to find the best active sampling approach to the assessment of occupational exposure to bioburden in Portuguese bakeries based on the data obtained with the use of specific impaction and impinger devices. We used impaction to collect fungal particles from 100 L air samples onto malt extract agar (MEA) supplemented with chloramphenicol (0.05 %). For growing fungi we also used dichloran glycerol (DG18) agar-based media and for mesophilic bacteria we used tryptic soy agar (TSA) supplemented with nystatin (0.2 %). For Enterobacteriaceae we used violet red bile agar (VRBA). With impingers we also collected 300 L air samples at the 300 L/min airflow rate, inoculated onto the same culture media. The two methods, impaction and impinger, showed statistically significant differences in the following counts: fungal on MEA (z=-2.721, p=0.007), fungal on DG18 (z=-4.830, p=0.000), total bacteria (z=-5.435, p=0.000), and Gram-negative coliforms (z=-3.716, p=0.000). In all cases the impaction method detected significantly higher concentrations than the impinger method. Fungal and bacterial loads were higher in the production unit and lower in the shop. The fungal load obtained with impaction varied between 10 and 5140 CFU m^-3, and total bacterial counts ranged between 10 and 4120 CFU m^-3. This study has shown that the impaction method is the best active sampling approach to assessing viable bioburden in this specific occupational environment, but a multi-faceted approach to sampling and analyses combining methods and media enables a more refined risk characterisation and, consequently, better tailored risk control measures to reduce adverse health outcomes in workers.