High-throughput 16S rRNA metagenomic sequencing has advanced our understanding of the gut microbiome, but its reliability depends on upstream processes such as DNA extraction and bacterial library preparation. In this study, we evaluated the impact of three different DNA extraction methods (a manual method with an ad hoc-designed pre-extraction phase (PE-QIA), and two automated magnetic bead-based methods (T180H and TAT132H)) and two bacterial library preparation protocols (home brew and VeriFi) on the 16S rRNA-based metagenomic profiling of faecal samples. T180H and TAT132H produced significantly higher DNA concentrations than PE-QIA, whereas TAT132H yielded DNA of lower purity compared to the others. In the taxonomic analysis, PE-QIA provided a balanced recovery of Gram-positive and Gram-negative bacteria, TAT132H was enriched in Gram-positive taxa, and T180H was enriched in Gram-negative taxa. An analysis of Microbial Community Standard (MOCK) samples showed that PE-QIA and T180H were more accurate than TAT132H. Finally, the VeriFi method yielded higher amplicon concentrations and sequence counts than the home brew protocol, despite the high level of chimeras. In conclusion, a robust performance in terms of DNA yield, purity, and taxonomic representation was obtained by PE-QIA and T180H. Furthermore, it was found that the impact of PCR-based steps on gut microbiota profiling can be minimized by an accurate bioinformatic pipeline.