There is a paucity of rigorous evidence on the impact of microcredit for the ultra-poor. Studies on microcredit do not tend to explore what, how and why different types of solutions work best for different types of clients. The available evidence gives a mixed picture on the impact of microcredit in general and for the ultra-poor in particular. The ultra-poor are commonly not reached by mainstream microcredit services. When they are reached, some studies find that microcredit can have positive impacts; others that it has disproportionately less of an impact and sometimes a regressive effect on the poorest. Experts warn that, given microcredit makes some people poorer and not richer, it is imperative to be particularly cautious when targeting the ultra-poor. Conventional microcredit models can be adapted to meet the needs of the ultra-poor, across the following areas: identifying and removing barriers to participation; monitoring; flexible loan conditions; supervision, training and support; linking economic strengthening with other services; and providing a ladder of financial services. Microfinance institutions (MFIs) face trade-offs between pursuing profit and reaching the poorest; some programmes serving the poorest may not be financially sustainable without additional support. A number of ‘graduation programmes’ aim to bridge traditional re-distribution schemes and conventional microfinance programmes. They help participants to graduate to using mainstream microfinance services, via the transfer of income-generating assets, food aid, health services and skills training. There is a growing evidence base on the impact of these types of programmes. This brief report is organised in four sections: 1. it summarises evidence on the impact of microcredit for the ultra-poor 2. it presents findings on supply-side design of microcredit for the ultra-poor 3. it focuses on apex funds for the ultra-poor 4. it summarises evidence on the approach to graduating the ultra-poor into microcredit.