Some ultra diffuse galaxies (UDGs) reveal many more globular clusters (GCs) than classical dwarf galaxies of the same stellar mass. These UDGs, with a mass in their GC system ($M_{\rm GC}$) approaching 10 per cent of their host galaxy stellar mass ($M_{\ast }$), are also inferred to have high halo mass to stellar mass ratios ($M_{\rm halo}/M_{\ast }$). They have been dubbed Failed Galaxies. It is unknown what role high GC formation efficiencies and/or low destruction rates play in determining the high $M_{\rm GC}/M_{\ast }$ ratios of some UDGs. Here we present a simple model, which is informed by recent JWST observations of lensed galaxies and by a simulation in the literature of GC mass loss and tidal disruption in dwarf galaxies. With this simple model, we aim to constrain the effects of GC efficiency/destruction on the observed GC richness of UDGs and their variation with the integrated stellar populations of UDGs. We assume no ongoing star formation (i.e. quenching at early times) and that the disrupted GCs contribute their stars to those of the host galaxy. We find that UDGs, with high $M_{\rm GC}/M_{\ast }$ ratios today, are most likely the result of very high GC formation efficiencies combined with modest rates of GC destruction. The current data loosely follow the model that ranges from the mean stellar population of classical dwarfs to that of metal-poor GCs as $M_{\rm GC}/M_{\ast }$ increases. As more data becomes available for UDGs, our simple model can be refined and tested further.