Most large-scale production processes in biotechnology are performed in fed-batch operational mode. In contrast, the screenings for microbial production strains are run in batch mode, which results in the microorganisms being subjected to different physiological conditions. This significantly affects strain selection. To demonstrate differences in ranking during strain selection depending on the operational mode, screenings were performed in batch and fed-batch modes. Two model populations of the methylotrophic yeast Hansenula polymorpha RB11 with vector pC10-FMD (P(FMD)-GFP) (220 clones) and vector pC10-MOX (P(MOX)-GFP) (224 clones) were applied. For fed-batch cultivations in deep-well microtiter plates, a controlled-release system made of silicone elastomer discs containing glucose was used. Three experimental set-ups were investigated: batch cultivation with (1) glucose as a substrate, which catabolite represses product formation, and (2) glycerol as a carbon source, which is partially repressing, respectively, and (3) fed-batch cultivation with glucose as a limiting substrate using the controlled-release system. These three experimental set-ups showed significant variations in green fluorescent protein (GFP) yield. Interestingly, screenings in fed-batch mode with glucose as a substrate resulted in the selection of yeast strains different from those cultivated in batch mode with glycerol or glucose. Ultimately, fed-batch screening is considerably better than screening in batch mode for fed-batch production processes with glucose as a carbon source.