Ambitious targets have been set for HIV prevention. We will begin by examining assumptions for HIV prevention scale-up in the light of the concept of the ‘prevention cascade’ using available data. Then, we will show that a key part of optimising prevention impact will be in allocating available funding according to four main factors -- intervention, population, geography and time. The impact of flexibility or constraint on each of these will be illustrated, providing recommendations for how international and domestic decisions can be taken to maximise epidemic impact. Finally, we will examine methods to evaluate the evidence of new prevention technologies to determine the scale of investment they might optimally attract. We will show that this requires a holistic view of the range of tools available in combination prevention and across the full prevention cascade, and that without that focus, sub-optimal decisions may be taken.