This paper presents a large-scale study of the phenomenon 'uncitedness'. A literature review indicates that uncitedness is related to at least three factors: Field, document type, and time. To explore these factors and their mutual influence further, and at much larger scale than previous studies, the paper focuses on seven subject areas (arts & humanities; social sciences; computer science; mathematics; engineering; medicine; physics & astronomy), seven document types (articles; reviews; notes; letters; conference papers; books; book chapters), and a 20-year publication window (1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015). Documents are searched in Scopus, and retrieved year-by-year, discipline-by-discipline, and for each individual document type (total: 29,472,184 documents; 7,508,741 uncited documents). The results show great variance in uncitedness ratios between subject areas and document types. This is probably caused by a somewhat tacitly agreed upon genre hierarchy existing in all subject areas, yet with important local traits and differences. The importance of the time-dimension is documented. Time to first citation varies a great deal between subject areas, and the uncitedness ratio is consequently shown to be quite sensitive to the length of citation windows.