Discovery of digital items by scholars and the public is highly dependent upon effective metadata to ensure inclusion and prioritization in search engines. Subject descriptions based on controlled vocabulary, such as Library of Congress subject headings (LCSH), are particularly useful to enhance discovery, but they may be expensive to provide. In this case study we compared the results of computer-assisted indexing with manual selection of subject terms as part of an effort to enhance findability of journal information at the issue level. Our results suggest that incorporating both computer and human processes provide the highest impact for discovery of distinctive digital collections.