Sensitive and accurate detection of food-borne pathogens is of critical importance for the food industry. One way to evaluate food quality control systems is to inoculate a known number of cells of a test microbe, then later determine their recovery by e.g. viable counts. This task is facilitated by using fluorescent derivatives, so that added cells can be distinguished from native microflora. My lab has been working with an industry partner to develop brightly-fluorescing and genetically-stable strains of Salmonella, E.coli, Cronobacter, and Listeria for this purpose, via chromosomal integration of a new superfolding GFP we have developed (free-use GFP, fuGFP), in association with better genus-specific control elements i.e. promoters and ribosome binding sites (RBS). Genetic engineering of Listeria has been especially challenging, due to low transformation frequencies, codon-usage differences, the need for unusual negative selection markers, and quirks of how their RBS are recognised. Here I will share some hard lessons we have learned about how subtle genetic differences between gram-negative and gram-positive bacteria can have major impacts on the phenotype of desired genetic constructs, and on the associated project timelines.