Background: Aphasiologists are motivated to select "functional", "relevant" and "useful" items for use in therapy; yet the field lacks discussion on what is meant by these terms and how to identify such items. Aims: The purpose of this article is to review the meaning of "functionally relevant" in the aphasia treatment literature and to specify challenges in identifying potentially relevant items for therapy. Main Contribution: This article shows that aphasiologists lack clear definitions, strategies and concrete tools to assist with identification of functionally relevant items for language therapy. Two main categories of functional vocabulary are defined-personally chosen vocabulary and generally frequent vocabulary. The review of the existing aphasia literature demonstrates the strengths and weaknesses of these approaches. Two critical points are raised related to selection of therapy items using data from language corpora. Firstly, it is paradoxical that aphasiologists often try to target the most common vocabulary in therapy but that language corpora have not been used to determine the identity of the most frequent words. Secondly, the analyses of the language corpora show that the most frequent spoken words represent a wide variety of word classes, such as adjectives, adverbs and pronouns. Yet, only a few treatment studies have targeted words other than concrete nouns and verbs. Conclusions: There is a need to use objective sources to identify and choose treatment targets. In addition, more therapy attempts should be directed to words other than the most concrete nouns and verbs. Use of frequency-based lists provides one way to identify and increase the number of items that are potentially relevant across people. Frequency-based vocabulary lists can also be used as a tool when asking people with aphasia or their significant others to identify items that they personally think should be targeted in therapy. Without creating and publishing common tools for this purpose, change in practice is difficult.