I was giving an introductory session to Art resources for a group of new postgraduates to help familiarise them with the printed collection (and its issues) and some of the online resources available. Over the course of my presentation I was surprised at my own assumptions of (perceived) knowledge of information retrieval, falling somewhere in the middle of the SCONUL Pillars grid which is one of several scales that map levels of information literacy. I was explaining information discovery by demonstrating a pivot search using Library of Congress subject headings. Use of databases and OPACs, as opposed to Google, operate on a few principles: some acquaintance with Boolean logic, knowledge of trucation and the principles of controlled vocabulary. I don’t expect everyone understands (or needs to learn) how fields index, the role of authority files, etc., but search strategies fundamentally built and informed by these principles will improve accuracy and progress information literacy development.In hindsight another surprise was that as penultimate presenter I thought I’d be duplicating what others had presented but in fact much of it was new or quasi-new to the participants. I think the lesson learned has less to do with holes in potential staff knowledge training and information induction and more to do with testing underlying assumptions and a fundamental return to basics; tabula rasa.