There is already substantial evidence to suggest that microfinance has polarising effects (Buckley, 1997, Copestake, 2002, Hulme, 2000 Marr, 2002; Wright, 2001). This should come as no surprise. Whilst every effort should be made to reduce negative impacts, it is inevitable that some clients will benefit more than others. Research needs to be designed in such a way as to reveal the reasons behind diverse impact. This paper draws upon a series of studies of village banking programmes in Peru to argue in favour of using a qualitative indepth individual interview protocol (QUIP) to reveal and analyse diverse impact. It finds that those least able to benefit were low-income women with dependent children. In addition, those clients exiting the programme did so due to an inability to repay loans due to their own ill health or that of their relatives, pointing to the importance of kinship obligations and the highly gendered dimension of health care which prevents them from using loans for their own businesses. Conflict within groups was found to be another major reason for exit.