It's just not that way. Program Directors, who seldom wear suits, push to get budgets for music tests because tests almost always help increase ratings. They look for more songs to add and songs that have not done well for a while that may have recovered a bit and for crossovers that "fit" in listener minds.
If a song tests positively, it is not burnt out and does not bore people. A music test is based on asking, for each song, "how much would you like to hear this song today?"
And if people don't want to hear it, why the heck would you want to play it? Testing is asking listeners to tell you what to play.
In any case, the reason why so few newer songs test is that the alt field is fragmented into multiple groups who don't care much for the well-liked songs of the other groups. So fining songs that all the possible listeners like is very hard because there are few of them.
That is because the "rock universe" is shrinking in a multi-ethnic rhythmic New York market.
And if they did this day after day, they would get a 0.1 share. There is a reason why they don't do it regularly. And, again, consultants and management don't pick the songs. The listeners do.