@B.Brekke It's not a "hope". If you look carefully, the standard perturbative expansion in loop-order is an expansion in the coupling constants. For coupling constants <<1, it's clear that the lower-order terms are the relevant ones.
Arguably it's even
better to not try to compute every term, since the perturbation series is
asymptotic and is
not expected to converge to the correct value.