Infectious gastroenteritis is a significant health issue globally. Identifying the causative pathogen is crucial for treatment, infection control and epidemiological surveillance. While PCR-based analyses are fast and sensitive, they only detect known pathogens. Clinical metagenomics can potentially identify novel or unexpected pathogens. This study aimed to evaluate shotgun metagenomics for detecting diarrhoeal pathogens in faecal samples from patients with infectious gastroenteritis and spiked samples from healthy donors, compared to PCR. DNA from clinical faecal samples (n = 12), spiked samples (n = 36), and control samples (n = 7) were analysed by PCR and shotgun metagenomics sequencing. Reads were taxonomically assigned, assembled, and binned into MAGs. MAGs were taxonomically assigned, and virulence genes were detected in bacterial assemblies and MAGs. Pathogens detected by PCR were also identified by taxonomic assignment of reads, though with lower sensitivity. Taxonomic assignment of MAGs identified 50% of bacterial pathogens and HAdV-F. Additional potential pathogens were observed in most samples. More bacterial virulence genes were detected in assemblies than in MAGs. In spiked samples, C. jejuni and HAdV-F were detected by both PCR and metagenomics, with significant correlation between Cq values and reads. Parasites were detected by few reads. Metagenomics has lower sensitivity compared to PCR but can provide supplementary information relevant for treatment. Challenges include additional potential pathogens, background microbiome, and introduced kitome, necessitating optimized extraction methods and strict quality controls.