Stroke affects 120 000 people each year in the UK and is the most common cause of adult-onset disability. Most stroke patients are cared for at home by informal carers. Support for these carers is provided by the community services, but although single-item measures have shown that there is a relatively high level of dissatisfaction with those services, there is no specific satisfaction measurement instrument. The present study aimed to construct and initially validate a questionnaire to measure carers' satisfaction with community services for clinical and research purposes. Qualitative interviews with a purposive sample of informal carers of stroke patients identified via stroke unit discharge records gave information about content, context and language. Together with information from the carers' literature, this formed the basis of the questionnaire. A self-completion questionnaire proved unacceptable to carers with a very poor response rate (20%). Face-to-face interviews were more acceptable, with the questionnaire taking approximately 10 minutes to administer. All carers approached agreed to take part and only three carers dropped out from follow up, all because of reasons not related to the study. Initial validity and reliability testing with a sample of 44 carers identified through stroke groups and general practice showed good correlation with a single-item satisfaction measure (RHO = 0.797), test–retest reliability (RHO = 0.885) and inter-rater reliability (RHO = 0.868), and a high degree of internal consistency (Cronbach's alpha = 0.859). Further validation with larger and more diverse groups of informal carers is needed before the questionnaire can be considered to be a robust and reliable tool. Factor analysis revealed seven factors: information about community support and involving the carer; amount, appropriateness and coordination of services; information about stroke; speed of change and concern about the carer; listening to the carer and being heard; problem management; and confidence in and accuracy of information. These factors have marked similarities to those found in the field of patient satisfaction with community services.