Taxonomic literature contains information about virtually ever known species on Earth. In many cases, all that is known about a taxon is contained in this kind of literature, particularly for the most diverse and understudied groups. Taxonomic publications in the aggregate have documented a vast amount of specimen data. Among other things, these data constitute evidence of the existence of a particular taxon within a spatial and temporal context. When knowledge about a particular taxonomic group is rudimentary, investigators motivated to contribute new knowledge can use legacy records to guide them in their search for new specimens in the field. However, these legacy data are in the form of unstructured text, making it difficult to extract and analyze without a human interpreter. Here, we used a combination of semi-automatic tools to extract and categorize specimen data from taxonomic literature of one family of ground spiders (Liocranidae). We tested the application of these data on fieldwork optimization, using the relative abundance of adult specimens reported in literature as a proxy to find the best times and places for collecting the species ( Teutamus politus) and its relatives ( Teutamus group, TG) within Southeast Asia. Based on these analyses we decided to collect in three provinces in Thailand during the months of June and August. With our approach, we were able to collect more specimens of T. politus (188 specimens, 95 adults) than all the previous records in literature combined (102 specimens). Our approach was also effective for sampling other representatives of the TG, yielding at least one representative of every TG genus previously reported for Thailand. In total, our samples contributed 231 specimens (134 adults) to the 351 specimens previously reported in the literature for this country. Our results exemplify one application of mined literature data that allows investigators to more efficiently allocate effort and resources for the study of neglected, endangered, or interesting taxa and geographic areas. Furthermore, the integrative workflow demonstrated here shares specimen data with global online resources like Plazi and GBIF, meaning that others can freely reuse these data and contribute to them in the future. The contributions of the present study represent an increase of more than 35% on the taxonomic coverage of the TG in GBIF based on the number of species. Also, our extracted data represents 72% of the occurrences now available through GBIF for the TG and more than 85% of occurrences of T. politus. Taxonomic literature is a key source of undigitized biodiversity data for taxonomic groups that are underrepresented in the current biodiversity data sphere. Mobilizing these data is key to understanding and protecting some of the less well-known domains of biodiversity.