NOTE: This poster won the second prize in the Poster Competition at the ZAW3 conference. The aim of this research was to establish what the best method was to generate a search query when attempting to locate academic publications through Google. Although some research outputs are available in university digital repositories, many users prefer using ordinary search engines to locate relevant information. This implies that these publications should be accessible through free-form Internet searching, over and above being accessible via the homepage of the repository. An online questionnaire was used to gather data on research outputs which were already published, and search queries were generated to locate the abstracts of these outputs. Based on prior research in the field, three different types of queries were used, including combinations of author surnames, keywords and the first sentence of the abstract. Journal articles, conference papers, book chapters and theses were investigated. Some abstracts did not rank amongst the top ten results, while others ranked well on Google. It was determined that the highest rankings were achieved when the first sentence of text was combined with author surnames. However, for an even higher success rate it is proposed that this method should be coupled with the second most successful one to produce a high degree of success.