In this chapter, I demonstrate how researchers can incorporate image content and emoji into corpus approaches to social media data. I present data from the Facebook Page of Dukki Gifts & Souvenirs, an independent business based in Nottingham, U.K. and, using a simple annotation scheme, I show how researchers can incorporate image content and emoji into frequency and collocation analysis in order to examine the patterns of use for images in this dataset. I discuss some of the adjustments required to extend approaches conventionally applied to written forms of data to image content, including how this might affect the way we think of core concepts, such as collocation.