High coupling oxide ferroelectrics are generally optimized through polarization rotation. Rather than the simple collinear piezoelectric coupling involving stretching of the polarization by applying an electric field or strain, larger coupling is obtained by applying an oblique field or shear strain that causes the polarization to change direction. The response is much softer for rotations of the polarization, so that much larger response can be achieved for the same field. This principal applies both to conventional piezoelectrics like PZT near the MPB and to relaxor ferroelectrics. What improvements can we look for from here, and what kind of studies are needed? Defect dipoles in oxides can greatly enhance electromechanical coupling when aligned. This is also due to the polarization rotation mechanism. Changes in bulk properties that can be unexpectedly large even for small concentrations of defects. We have simulated ferroelectric perovskites such as BaTiO3 as functions of defect dipole concentration and orientations. We have also performed experiments on Mn-doped samples as functions of aging. In an aging process, the defects align spontaneously, whereas in applied fields this process can be better controlled and accelerated. Now we have a detailed microscopic understanding of the effects of defect concentration, distribution and defect dipole alignment on polarization and strain response. This work is supported by the US Office of Naval Research and the European Research Council Advanced Grant ToMCaT.