That sounds to be a better idea, but people say (and as an outsider, it does sound believable, I guess) that that's pretty much impossible - see for example [1]:

> I regret to say that in reviewing perhaps a hundred grants or job applications and trying to find the ten grants to fund or one person to employ, I do not read every paper in the bibliography and assess the research on the basis of my limited understanding. I just don’t have the time or expertise to read and judge all the papers.

[1] https://theconversation.com/why-i-disagree-with-nobel-laurea...