We present new constraints on the Hubble function H(phi) and subsequently on the inflationary scalar potential V(phi) from WMAP 3-year data combined with the Sloan Luminous Red Galaxy survey (SDSS-LRG), using a new methodology which appears to be more generic, conservative and model-independent than in most of the recent literature, since it depends neither on the slow-roll approximation, nor on any extrapolation scheme for the potential beyond the observable e-fold range, nor on additional assumptions about initial conditions for the inflaton velocity. Besides these new constraints, we will briefly discuss the accuracy of the slow-roll approximation in the light of present day observations, and the constraining power of extrapolation, which are both widely used in other works.