The analysis of microbiota composition in humans, animals, and built environments is important because of emerging roles and applications in a broad range of disease and ecological phenotypes. Next Generation Sequencing is the current method of choice to characterize microbial community composition. The taxonomic profile of a microbial community can be obtained either by shotgun analysis of random DNA fragments or through 16S ribosomal RNA gene (rDNA) amplicon sequencing. It has been previously shown that the 16S rDNA amplicon sequencing approach yields quantitatively and qualitatively different results compared to shotgun metagenomics when the two techniques are used to assess microbial community composition on the same samples. However, most of such comparisons were either based on the recovery of 16S rDNA sequences in the shotgun metagenomics data or limited to a single microbiome or synthetic samples. Direct comparison of shotgun metagenomics and 16S rDNA amplicon sequencing on the same samples was performed only once in the recent literature, suggesting that the two methods yield comparable results. Here, we set out to compare the outcome of these two alternative approaches to the microbiome characterization in human gut microbiomes from stool samples. To this end, we processed six different samples with both techniques. We report here that shotgun next generation sequencing metagenomics allows much deeper characterization of the microbiome complexity, allowing identification of a larger number of species for each sample, compared to 16S rDNA amplicon sequencing. Further comparative studies in independent samples are called for.