Classic taxonomy is a powerful tool for identifying animals based on morphology but has shown to be problematic on similar looking, cryptic species. A solution to this problem has been found within the bauplan of life, the DNA (deoxyribonucleic acid). DNA is used to create and regulate proteins. The structure of DNA has highly unique sections that are conserved within species, but diverse between species. One particular section, a 648 bp long fragment of the mitochondrial cytochrome c oxidase subunit 1 (CO1) gene, has become a popular barcode for species identification. Here, a new barcoding technique, character-based barcoding more similar to traditional approaches is tested. This thesis investigates whether CO1 is suitable as a single marker (a) or should be complemented by others (b). Performance of distance- and character-based barcoding (c) is evaluated and it is tested whether character-based barcoding can be used to identify cryptic species (d). In the first manuscript, CO1 sequences of endangered turtle species are compared (a). Having a reliable tool for species identification is an important asset in species protection surveillance. Variability within the barcode region is assessed and the utility of both distance- and character-based methods for species identification are evaluated (c). Odonata is an old order rich in species. As many species have evolved in a short time, it was observed that intra- and interspecific variety is overlapping in some sister groups. This observation made Odonata the ideal candidate for testing CO1 (a), ND1 (b), as well as distance- and character-based-barcoding (c) in the second manuscript. Ants are prime examples for high degrees of cryptic biodiversity due to complex population differentiation, hybridization and speciation processes. As combinations of multiple marker regions seemed to be a better approach to barcoding, three markers (CO1, 28S rDNA, rhodopsin) are tested (b) in the third manuscript. A combined, layered approach to character-based barcoding is evaluated and unique diagnostics specific to geolocations are identified (d). The results of all three studies show that combining multiple markers improves identification success. The character-based approach provides better identification in the tested animal groups. This method can be used to estimate presence, absence or frequency of cryptic species.