Timely diagnosis of human immunodeficiency virus (HIV) enables access to antiretroviral treatment, which reduces mortality, morbidity and further transmission in people living with HIV. In the UK, late diagnosis among black African people persists. Novel methods to enhance HIV testing in this population are needed. To develop a self-sampling kit (SSK) intervention to increase HIV testing among black Africans, using existing community and health-care settings (stage 1) and to assess the feasibility for a Phase III evaluation (stage 2). A two-stage, mixed-methods design. Stage 1 involved a systematic literature review, focus groups and interviews with key stakeholders and black Africans. Data obtained provided the theoretical base for intervention development and operationalisation. Stage 2 was a prospective, non-randomised study of a provider-initiated, HIV SSK distribution intervention targeted at black Africans. The intervention was assessed for cost-effectiveness. A process evaluation explored feasibility, acceptability and fidelity. Twelve general practices and three community settings in London. HIV SSK return rate. Stage 1 - the systematic review revealed support for HIV SSKs, but with scant evidence on their use and clinical effectiveness among black Africans. Although the qualitative findings supported SSK distribution in settings already used by black Africans, concerns were raised about the complexity of the SSK and the acceptability of targeting. These findings were used to develop a theoretically informed intervention. Stage 2 - of the 349 eligible people approached, 125 (35.8%) agreed to participate. Data from 119 were included in the analysis; 54.5% (65/119) of those who took a kit returned a sample; 83.1% of tests returned were HIV negative; and 16.9% were not processed, because of insufficient samples. Process evaluation showed the time pressures of the research process to be a significant barrier to feasibility. Other major barriers were difficulties with the SSK itself and ethnic targeting in general practice settings. The convenience and privacy associated with the SSK were described as beneficial aspects, and those who used the kit mostly found the intervention to be acceptable. Research governance delays prevented implementation in Glasgow. Owing to the study failing to recruit adequate numbers (the intended sample was 1200 participants), we were unable to evaluate the clinical effectiveness of SSKs in increasing HIV testing in black African people. No samples were reactive, so we were unable to assess pathways to confirmatory testing and linkage to care. Our findings indicate that, although aspects of the intervention were acceptable, ethnic targeting and the SSK itself were problematic, and scale-up of the intervention to a Phase III trial was not feasible. The preliminary economic model suggests that, for the acceptance rate and test return seen in the trial, the SSK is potentially a cost-effective way to identify new infections of HIV. Sexual and public health services are increasingly utilising self-sampling technologies. However, alternative, user-friendly SSKs that meet user and provider preferences and UK regulatory requirements are needed, and additional research is required to understand clinical effectiveness and cost-effectiveness for black African communities. This study is registered as PROSPERO CRD42014010698 and Integrated Research Application System project identification 184223. The National Institute for Health Research Health Technology Assessment programme and the BHA for Equality in Health and Social Care.