There is extensive genetic substructure within the species Escherichia coli. In 2000 a simple triplex PCR method was described by Clermont and colleagues that enables an E. coli isolate to be assigned to one of the phylo-groups A, B1, B2 or D. The growing body of multi-locus sequence data and genome data for E. coli has refined our understanding of E. coli's phylo-group structure and eight phylo-groups are now recognized: seven (A, B1, B2, C, D, E, F) belong to E. coli sensu stricto, whereas the eighth is the Escherichia cryptic clade I. Here a new PCR-based method is developed that enables an E. coli isolate to be assigned to one of the eight phylo-groups and which allows isolates that are members of the other cryptic clades (II to V) of Escherichia to be identified. The development of the method is described and the method is validated. Over 95% of E. coli isolates can be correctly assigned to a phylo-group. Two collections of human faecal isolates were screened using the new phylo-group assignment method demonstrating that about 13% of E. coli isolates belong to the newly described phylo-groups C, E, F and clade I.