And what we learn from evolution is we should laugh at should too. My theory: should is a gambit to control you on the part of other people trying to win in the game of evolution.
Thus, a woman benefits by selecting a loyal man rather than a promiscuous one. For her to marry a loyal man with strong genetics (e.g. an attractive man with a limited sexual history, indicating a low disposition toward promiscuity) is optimal. A woman who disqualifies men because they've had a low number of sexual partners is never going to accomplish this.
Anyway, the obsession over the number of past sexual partners means nothing in an evolutionary context, since the whole concept is a social construct. Pre-numerate cave dwellers would have lost count pretty quickly.
Anything's reproductive "goal", in any times, is to show up a lot in later generations. Promiscuity and particularity both show up as strategies in humans and broadly beyond. Anybody studying reporductive strategies many generations later will have a strong survivorship bias in their results.
A woman who selects a man who isn't very promiscuous is unlikely to get her genes spread very far by her son(s). This trades off against the part of the story you tell.