Background & Aims A recent systematic review of conversation training for communication partners of people with aphasia has shown that it is effective, and improves participation in conversation for people with chronic aphasia. Other research suggests that people with aphasia are better able to learn communication strategies in an environment which closely mirrors that of expected use, and that cognitive flexibility may be a better predictor of response to therapy than severity of language impairment. This study reports results for a single case, one of a case series evaluation of a programme of conversation training for agrammatism that directly involves a person with aphasia (PWA) as well as their communication partner. It explores how a PWA is able to engage with and learn from the therapy, and whether this leads to qualitative change in post-therapy conversation behaviours. Methods & Procedures A 55-year-old man with chronic agrammatism and his wife took part in eight weekly sessions of conversation therapy, adapted from Supporting Partners of People With Aphasia in Relationships and Conversation (SPPARC). Language and conversation were assessed before and after therapy, and the couple's views on conversation and disability were elicited. Conversation analysis was used to analyse: (1) pre-therapy conversation patterns, (2) how the PWA engaged and learned during therapy and the forms of facilitation that aided this process, and (3) qualitative change in post-therapy conversation behaviour. Outcomes & Results After therapy, the PWA showed increased insight and acceptance of the use of strategies such as writing and drawing in the face of conversational difficulty. However, use was prompted by his wife and was rarely spontaneous. Conclusions & Implications This single case study suggests that conversation training based around an experiential learning process is able to engage a PWA directly in learning about the effects of aphasia on conversation. Key facilitators were self-study via video and experience of practising conversation whilst receiving online feedback from a speech and language therapist. However, increased insight did not automatically change conversation behaviour. Although he better understood the effects of his aphasia on conversations with his wife, learning stopped short of the ultimate goal of the conversation training programme; the spontaneous use of strategies worked on in therapy when faced with conversation breakdown. One explanation may be that limited cognitive flexibility lead to problems with switching from one strategy to another.