A number of culture-independent DNA-based methods are commonly used to determine the phylogenetic affiliation of microbes in a complex community such as sponges which termed as culture independent method. These are considered to be more accurate as a larger number of genera and species are observed compared to culture-based methods. However, different molecular methods often give different results with the same samples. T-RFLP analysis has been viewed as a rapid, convenient, sensitive, high-throughput, and highly reproducible method in characterising strains to the genus level in microbial ecology studies. Mass sequencing technologies such as pyrosequencing have introduced a level of accuracy in characterising microbial diversity, often to the species, though in terms of cost and analysis it is still expensive. There are usually discrepancies between results obtained from TRFLP analysis and pyrosequencing and determining the causal factor(s) is of paramount importance. A detailed analysis showed that the primers used in both techniques must be able to “capture” the same complete compliment of genera/species, even though the primer set is not the same as different product lengths are required for each analysis. This was achieved by using two newly designed universal primers which has been developed in our lab. The TRFLP analysis with these newly designed primer sets, and the three most suitable restriction digestion enzymes Hha1, Msp1 and Rsa1 showed that one of the primer sets had a 95% overlap when compared with pyrosequencing data.