Background: While evidence on the use of semantic therapy to improve word retrieval has increased (Whitworth, Webster, & Howard, 2005), the impact of semantic therapy on comprehension itself has not been studied in depth. This paper reports on a specific semantic therapy, which aims to improve access to semantic information. Data from two participants, JAC and AD, are presented. Aims: The study aimed to examine the effectiveness of a specific semantic therapy on comprehension of treated and untreated items, and on related tasks, thereby examining generalisation of any improvement to both items and tasks involving semantic processing. Methods & Procedures: The participants’ comprehension was assessed in detail prior to therapy. Measures were repeated over time to ensure any improvement seen could not be due to more general recovery mechanisms. The two participants then took part in therapy which aimed to improve their comprehension, using a picture name verification task. Following therapy, relevant assessments were repeated to examine the effects of therapy. The study uses single case methodology with comparison across two similar cases. Outcomes & Results: JAC and AD responded differently to the semantic therapy; JAC’s auditory comprehension performance improved and this was regardless of whether items had been treated or not. In contrast, despite having a broadly similar profile of language ability, AD’s performance did not change. Potential reasons for this are explored within the paper. Conclusions: This paper provides a much needed demonstration of semantic therapy for comprehension problems. One specific therapy is tested, rather than the combined therapies used in the existing literature. It considers the extent to which therapy generalises to untreated words. It is clear that further research is needed regarding finer differential diagnosis, and ultimately, the impact of therapy on everyday comprehension for these individuals.