A cross-sectional survey of Giardia duodenalis and Cryptosporidium parvum infection in the water buffalo (Bubalus bubalis) was carried out in central Italy. A geographical information system (GIS) was constructed utilizing as data-layers the topographic base map and the digital aerial photographs of the study area, as well as the geo-referenced points of all the buffalo farms. The survey was conducted on a sample of 90 farms, selected using a grid approach followed by proportional allocation. For this purpose, a grid representing quadrants of 5 x 5 km was overlaid on the study area within the GIS. As a result, the study area was divided in equal quadrants, and the number of farms sampled in each quadrant was proportional to the total number of study population in that quadrant. On each farm, faecal samples were collected per rectum from three to five asymptomatic buffalo calves, aged from 1 to 9 weeks. The total number of faecal samples collected was 347. Each faecal sample was tested for the presence of copro-antigens of G. duodenalis and of C. parvum using two commercially available enzyme-linked immunosorbent assays. Out of the 90 farms, 27 (30.0%) resulted positive for G. duodenalis and 22 (24.4%) for C. parvum. Co-infection was found in ten (11.1%) farms. With respect to animals, out of the 347 faecal samples, 63 (18.1%) were found to have antigens of G. duodenalis and 51 (14.7%) of C. parvum. Co-infection was found in ten buffalo calves (2.9%). The results of the logistic regression models showed a positive association between the positivity to G. duodenalis and the presence of sheep on farm and between the positivity to C. parvum and the high number of buffaloes on farms. No significant co-infection between the two protozoa was found. In conclusion, the findings of the present study, derived from a systematic territorial survey planned with GIS, are noteworthy because they provided additional data on C. parvum and the first evidence of G. duodenalis infection in water buffaloes.