The 16S rRNA gene is one of the most commonly used molecular markers for estimating bacterial diversity during the past decades. However, there is no consistency about the sequencing depth (from thousand to millions of sequences per sample), and the clustering methods used to generate OTUs may also be different among studies. These inconsistent premises make effective comparisons among studies difficult or unreliable. This study aims to examine the necessary sequencing depth and clustering method that would be needed to ensure a stable diversity patterns for studying fish gut microbiota. A total number of 42 samples dataset of Siniperca chuatsi (carnivorous fish) gut microbiota were used to test how the sequencing depth and clustering may affect the alpha and beta diversity patterns of fish intestinal microbiota. Interestingly, we found that the sequencing depth (resampling 1000-11,000 per sample) and the clustering methods (UPARSE and UCLUST) did not bias the estimates of the diversity patterns during the fish development from larva to adult. Although we should acknowledge that a suitable sequencing depth may differ case by case, our finding indicates that a shallow sequencing such as 1000 sequences per sample may be also enough to reflect the general diversity patterns of fish gut microbiota. However, we have shown in the present study that strict pre-processing of the original sequences is required to ensure reliable results. This study provides evidences to help making a strong scientific choice of the sequencing depth and clustering method for future studies on fish gut microbiota patterns, but at the same time reducing as much as possible the costs related to the analysis.