A new proposal for taxonomic species description is presented to replace the traditional descriptive texts. This is an attempt to enhance the species description rate and to make the description output available to other scientific disciplines, machine learning, lucid identification keys, big data analysis and its applications. The method consists in presenting the description of the overall morphology in a coded matrix, following a character list with detailed observed conditions for each character. The method is supposed to be dynamic and open to amendments and new data addition as they become available. We test the new method describing five new species of Collembola Symphypleona of the genus Pararrhopalites as a generalized model and made the coded output available. We conclude that a coded taxonomic description is an advance to the traditional taxonomic text, with potential to enhance the global descriptions rate. The generated data is a dynamic matrix that can be expanded with any data that becomes available, also it can be easily used in other fields of science, allowing non-experts to access the data for phylogenetic, biogeographic, ecological studies and big data analysis. Furthermore, it is a step forward to a general template to semi-automated taxon recognition and auxiliary tools for species description using machine learning.