You should consider costs. This tradeoff is a large part of what the field of epidemiology studies. Turns out vaccines don't need to be
that effective to be worthwhile, because of the way disease spreads. There's this concept of a "reproduction number" (r-naught) that predicts how many people a given infected person will pass the disease to, based on a combination of the virulence of a disease, how long infected people remain contagious, how much contact people have with each other, etc. The flu often has a value of around 2, meaning the average infected person will infect two others before they're over it.
You can use this number to estimate how fast a disease will spread, how likely it is to be successfully contained, or in the case of seasonal diseases, how many people will get it before seasonal societal changes (school getting out, people spending more time outside, etc.) will effectively curtail transmission for the year.
Looking at a vaccine on an individual basis, if it's only, say, 50% effective, that sounds pretty awful -- maybe almost not worth the bother. But on a society-wide basis, if every person were vaccinated, that would effectively knock the R0 value from 2 to 1, which is huge. Transmission is exponential: if you infect two people and they infect two people and they infect two people, etc., the total number of infected looks dramatically different than if everybody only infects one person. So in isolation it's only 50%, but societally, with high vaccination rates, there's a huge multiplicative effect that has the potential to dramatically reduce each person's risk of infection if vaccination rates are high enough. And that's still true even if it's even less effective than 50%, and has all sorts of economic and social consequences.