BACKGROUND: Conversational coaching involves teaching communication strategies to individuals with aphasia and their spouses. Strategies are chosen by the couple and are taught and practised in the context of a conversation. Although strategy training has been explored in previous research studies, many questions regarding treatment implementation and outcomes remain unanswered. AIMS: The aims of this study were to investigate the effects of conversational coaching and to determine variables for consideration in future efficacy research of this treatment technique. METHODS AND PROCEDURES: Two couples participated in a single-subject experimental design across subjects. Mr and Mrs Y were 76 and 70 years old, and Mr and Mrs G were 41 and 39 years old, respectively. During baseline sessions, the individual with aphasia watched a videotaped story about a real-life event and then attempted to share the content of that story with his spouse. During treatment sessions, the same procedure was used, but the clinician intervened and coached both participants in the use of selected verbal and non-verbal strategies to improve the quality of the conversation. The primary dependent measure in the study was the number of main concepts successfully communicated during conversations. Outcome measures included pre/post treatment probes, standardised testing using the Communications Activities of Daily Living - Second Edition (CADL-2; Holland, Frattali, & Fromm, 1999) and social validation ratings. OUTCOMES AND RESULTS: Experimental data were variable for both dyads. However, positive outcomes were obtained. First, the percentage of main concepts successfully communicated was significantly greater during post-treatment probes than during pre-treatment probes for both dyads. Second, Mr Y demonstrated significant improvement in his total CADL-2 scores following treatment. Third, individuals who judged the quality of pre- and post-treatment conversations understood more of the conversation between both couples during post-treatment probe conversations than during pre-treatment ones. CONCLUSION: Positive outcomes, including participants' perceptions of treatment effects, support further experimental study of this technique. Strategy selection, stimuli for conversational topics, and procedural specificity of the intervention were variables identified as necessitating further research in a controlled experiment design.