Species identification from hair has been performed in the past by several techniques, such as scanning electron microscopy and polymerase chain reaction. Despite the great promise of mass spectrometry herein, the repetitive glycine stretches and the evolutionary conserved sequences of keratins make the results from conventional database search algorithms on MSMS fragmentation data very ambiguous. Here, we present a new method based on electron spray quadrupole time-of-flight (ESI-Q-TOF) mass spectrometry and spectral library searching. By comparing different sets of data processing parameters, spectral libraries for human, cat, and dog were constructed with the highest possible specificity and sensitivity. This proof of principle was confirmed by the annotation of blind samples. In addition, by providing a step-by-step roadmap for creating such libraries, more species can be included in the future as demonstrated here by the inclusion of sheep and rabbit. Additionally, we illustrate that this approach allows for species identification of a single hair, making this an interesting approach in a forensic setting.