The genus Kentisuchus was constructed by Charles Mook in 1955 as a replacement name for the basal tomistomine "Crocodylus" spenceri, named by William Buckland in 1836. Buckland named "Crocodylus" spenceri on the basis of a partial skull found from the Isle of Sheppey in Kent, England. Another tomistomine known as "Crocodylus" toliapicus was later described by Richard Owen, who considered it synonymous with “Crocodylus” spenceri yet kept the new name. In 1888 Richard Lydekker considered "C." toliapicus synonymous with two other known crocodilians, "C." champsoides, named by Owen, and "C." arduini, named by De Zigno, and reapplied the name "C." spenceri to all of these species.
The genus name Kentisuchus was constructed only after it was realized that these tomistomine specimens were clearly distinct from the genus Crocodylus and that some specimens originally assigned to "C." spenceri belonged to entirely different genera and species. "C." arduini was reassigned to the new genus Megadontosuchus in the same paper that Kentisuchus was first described in. "C." toliapicus and "C." champsoides were both found to be species of Kentisuchus, the former being considered the type species. "C." spenceri has provisionally been referred to Kentisuchus as well.
K. spenceri may be congeneric with M. arduini and Dollosuchus dixoni, according to several more recent studies. If this is the case then the name Dollosuchus has seniority over the other two, although it is likely the three taxon will remain their own distinct species. Other recently conducted phylogenetic analyses suggest that K. spenceri is closely related to Dollosuchus. An apparent close relationship between K. spenceri and Eosuchus lerichei has been used to imply that the latter species was a tomistomine, while it is now thought that Eosuchus is a basal gavialoid that is crownward to most other members of the superfamily.