We investigate the minimal performance, in terms of beam luminosity and detector size, of a neutrino factory to achieve a competitive physics reach for the determination of the mass hierarchy and the discovery of leptonic CP violation. We find that a low luminosity of 2 × 10 20 useful muon decays per year and 5 GeV muon energy aimed at a 10 kton magnetized liquid argon detector placed at 1300 km from the source provides a good starting point. This result relies on θ13 being large and assumes that the so-called platinum channel can be used effectively. We find that such a minimal facility would perform significantly better than phase I of the LBNE project and thus could constitute a reasonable step towards a full neutrino factory. The recent discovery of θ 13 [1-3] is a major step towards the completion of the leptonic mixing matrix. The remaining unknown mixing parameters, within a three neutrino framework, are the Dirac CP-violating phase, δ, and the ordering of the neutrino mass eigenstates, sgn(∆m 2 31 ). CP violation (CPV) within the Standard Model has proved to be quite intruiging in the hadronic sector already: even though the strong interaction seems to be conserving CP, it is significantly violated in quark mixing. Neutrinos now offer the third opportunity to learn more about the role of the CP symmetry in Nature. Also, if one considers the question of unitarity and the completeness of the three neutrino picture, the determination of the CP phase will play a crucial role, like it did in the quark sector.Direct CPV in neutrino oscillations can only be observed in appearance experiments, where the initial and final neutrino flavors are different. For practical reasons, this requirement confines experiments to study ν e ↔ ν µ andν e ↔ν µ transitions. Conventional neutrino beams are obtained from the decay of relativistic pions and therefore predominantly consist of ν µ orν µ , depending on whether π + or π Therefore, a number of new experiments has been proposed in order to observe CPV in the leptonic sector, see for instance Ref.[5]; in the U.S. context, this proposal is the long baseline neutrino experiment (LBNE). The first stage of the LBNE project comprises a 700 kW proton beam to produce pions and a 10 kton liquid argon (LAr) detector placed at a distance L = 1300 km from the source [6]. The CPV discovery potential is limited due to a lack of statistics, though. An upgraded beam in the multi-MW range (superbeam) would obviously yield a much better physics potential [7]. However, these beams are eventually limited by intrinsic backgrounds and systematic effects: large flux uncertainties, combined with the inability to measure the final flavor cross sections at the near detector, introduce large systematical errors which are very difficult to control [8,9].For the determination of the CP phase a similar precision to that achieved in the quark sector is only offered by a neutrino factory (NF) [9,10]. In a NF a highly collimated beam of muon neutrinos and electron antineutrinos is produced from muon decay...