Classically, the diagnosis of adenomyosis relied on histological examination of uteri following hysterectomy and classifications focused on the depth of endometrial invasion within the myometrium. There remain uncertainties around the cut-off point for the histological diagnosis. Imaging-based diagnosis enables recognition of the condition in women not undergoing surgery and facilitates the assessment of the extent of adenomyosis within the whole uterus, as well as of affections of the uterovesical pouch and of the pouch of Douglas. In this article, we explore the diagnostic uncertainties, the need to produce a classification of the condition and the challenges towards that goal. A distinction should be drawn between disease mapping and a classification that may link histological or image-based features with clinical characteristics, or with pathophysiology. An agreed system for reporting adenomyotic lesions may enable comparisons of research studies and thus contribute towards an informed classification. To this aim, we outline the features of the condition and explore the characteristics that are considered when producing a taxonomy. These include the latest proposal for subdivision of adenomyosis into an internal and an external variant. We also explore the uncertainties linked to classifying involvement of the uterovesical pouch, the pouch of Douglas and lesions in the outer myometrium. The limitations of currently available evidence suggest that agreement on a hypothesis to underpin a classification is unlikely at present. Therefore, current efforts will probably remain focused on disease mapping.