We contrast two approaches to calculating trans-Planckian corrections to the inflationary perturbation spectrum: the new physics hypersurface (NPH) model, in which modes are normalized when their physical wavelength first exceeds a critical value, and the boundary effective field theory (BEFT) approach, where the initial conditions for all modes are set at the same time, and modified by higher-dimensional operators enumerated via an effective field theory calculation. We show that these two approaches—as currently implemented—lead to radically different expectations for the trans-Planckian corrections to the cosmic microwave background (CMB) and emphasize that in the BEFT formalism we expect the perturbation spectrum to be dominated by quantum gravity corrections for all scales shorter than some critical value. Conversely, in the NPH case the quantum effects only dominate the longest modes that are typically much larger than the present horizon size. Furthermore, the onset of the breakdown in the standard inflationary perturbation calculation predicted by the BEFT formalism is likely to be associated with a feature in the perturbation spectrum, and we discuss the observational signatures of this feature in both CMB and large-scale structure observations. Finally, we discuss possible modifications to both calculational frameworks that would resolve the contradictions identified here.