A graphical abstract (GA) represents a piece of artwork that is intended to summarize the main findings of an article for readers at a single glance. Many publishers currently encourage authors to supplement their articles with GAs, in the hope that such a convenient visual summary will facilitate readers with a clearer outline of papers that are of interest and will result in improved overall visibility of the respective publication. To test this assumption, we statistically compared publications with or without GA published in Molecules between March 2014 and March 2015 with regard to several output parameters reflecting visibility. Contrary to our expectations, manuscripts published without GA performed significantly better in terms of PDF downloads, abstract views, and total citations than manuscripts with GA. To the best of our knowledge, this is the first empirical study on the effectiveness of GA for attracting attention to scientific publications. 1. Background In the field of publishing, where nowadays the majority of articles are searched for and read online and where scientists increasingly make use of social media platforms to share their findings, many publishers currently encourage authors to prepare a graphical abstract (GA) in addition to the written abstract. The GA is intended to summarize the article´s main findings for readers at a single glance, to attract audience attention, and to make readers pick out one´s article from a plethora of potentially interesting literature. It is also a format that is perfectly suitable for sharing via social media. However, to achieve effectiveness it is optimal to provide a well-designed and highly informative graph. Frequently, authors who do not want to put extra work into a GA simply compile some of the most important figures from their manuscript, a strategy that does not always lead to meaningful GA . It also has to be considered that GA is not equally suitable for all kinds of manuscripts.