Dogs are important hosts and reservoirs of leishmaniasis, a disease caused by protozoan parasites from the genus Leishmania, affecting ~12 million people worldwide. The detection of visceral leishmaniasis (VL) in dogs by real-time PCR (qPCR) may improve on diagnosis, but the different qPCR methods available for Leishmania DNA detection have not been established as routine in diagnostic tools and/or epidemiologic studies for canine VL. Here, we compared three qPCR assays (DNApol, Linj31, and LDON) in the detection of VL by Leishmania infantum in spleen (n = 48; 7), skin (n = 48; 7), and whole blood (n = 44; 7) samples from serologically positive and negative dogs, respectively. Overall, the DNApol performed better than the Linj31 and LDON assays in the detection of positive samples in all tissues tested, yielding from 66.7 to 100.0% of positivity for both skin and spleen samples. For spleen samples, we observed no statistically significant differences between positive detection by the LDON and DNApol assays. Whole blood samples yielded the lowest rates of positive detection, regardless of the qPCR assay used. In contrast, positive detection of Leishmania DNA was as efficient from skin samples using the DNApol assay as from spleen samples using either the DNApol or the LDON assay. Although qPCR assays from skin samples may not be practical for use in the field, our study suggests that the DNApol and LDON assays from skin samples could be used in future to evaluate canine VL treatment in veterinary clinics.