DNA extraction yield from drinking water distribution systems and premise plumbing is a key metric for any downstream analysis such as 16S amplicon or metagenomics sequencing. This research aimed to optimize DNA yield from low-biomass (chlorinated) reverse osmosis-produced tap water by evaluating the impact of different factors during the DNA extraction procedure. The factors examined are (1) the impact of membrane materials and their pore sizes; (2) the impact of different cell densities; and (3) an alternative method for enhancing DNA yield via incubation (no nutrient spiking). DNA from a one-liter sampling volume of RO tap water with varying bacterial cell densities was extracted with five different filter membranes (mixed ester cellulose 0.2 μm, polycarbonate 0.2 μm, polyethersulfone 0.2 and 0.1 μm, polyvinylidene fluoride 0.1 μm) for biomass filtration. Our results show that (i) smaller membrane pore size solely did not increase the DNA yield of low-biomass RO tap water; (ii) the DNA yield was proportional to the cell density and substantially dependent on the filter membrane properties (i.e., the membrane materials and their pore sizes); (iii) by using our optimized DNA extraction protocol, we found that polycarbonate filter membrane with 0.2 μm pore size markedly outperformed in terms of quantity (DNA yield) and quality (background level of 16S gene copy number) of recovered microbial DNA; and finally, (iv) for one-liter sampling volume, incubation strategy enhanced the DNA yield and enabled accurate identification of the core members (i.e., and as the most abundant indicator taxa) of the bacterial community in low-biomass RO tap water. Importantly, incorporating multiple controls is crucial to distinguish between contaminant/artefactual and true taxa in amplicon sequencing studies of low-biomass RO tap water.