Background: In recent years conversation has become an area of interest for aphasia therapy, with several studies using conversation analysis (CA) to target and evaluate therapy. Most of these studies have focused on the main conversation partner of the person with aphasia, and in particular have targeted the partner's pedagogic behaviours in relation to the person with aphasia. Evaluations of therapy have primarily taken the form of qualitative analyses of change in conversational behaviours. Aims: This single-case intervention study aims to advance research into interaction-focused intervention for aphasia in the following ways: by targeting intervention at the person with aphasia and the main conversation partner as a couple; by focusing on conversational behaviours where the person with aphasia can be seen to be restricted by the conversational actions of the conversation partner, in particular by recurrent questioning using closed questions and yes/no interrogatives; and by using a novel combination of qualitative and quantitative approaches to evaluate the intervention. Methods & Procedures: CA was used to target and evaluate interaction-focused intervention for a couple where one partner has aphasia. Evidence for change was evaluated using qualitative and quantitative evidence of change in conversational behaviours; evidence from naïve raters of pre- and post-intervention conversation extracts; and interview/other feedback from the conversation partner. Outcomes & Results: There was evidence that the intervention had changed the couple's conversational behaviours. In particular, the conversational behaviours of the non-aphasic partner were in general less restricting for the person with aphasia in that she was now using fewer questions and more instance of other types of turns, such as paraphrases. Following intervention the person with aphasia had also changed in that he was now producing turns that had more sentences, or attempts at sentences, and which developed the topic of talk across several of his turns. Conclusions: The study provides evidence that directly targeting the conversational behaviours of the person with aphasia and/or a main conversational partner can produce positive change, and can achieve this in a way that is ecologically valid. In particular, it highlights the usefulness of targeting conversational behaviours that are proving to be maladaptive for the participants. It provides further evidence that creating change in the non-aphasic partner's conversational behaviour may facilitate change in the person with aphasia's conversational and linguistic performance. © 2010 Psychology Press.