When does a high-impact study make its impact?

The epithome of a high-impact article in my field is Carolyn Miller's "Genre as Social Action" from 1984. It is to a wide extent the center piece of genre research. It is so influential that you cannot simply insert it into an article as "Miller (1984)". You have to call it something along the line of "Miller's groundbreaking/trailblazing/influential/seminal 1984-study". 

You don't write that just to pay homage (though homage is indeed due), but more to demonstrate that you know the field well enough to be aware that this is the study you have to pay homage to. It's sort of a rite of passage. 

Moreover, with too many texts to read and too little time to read them, it is also an excellent shibboleth to protect your time. If a text on genre does not have Miller on its bibliography, it is well-nigh certain that it's not worth your time to read. Miller may be replaced by a solid line-up of other central genre researchers, Freadman, Devitt, Bakhtin, Bazerman, Schryer, and others (see, for instance, this list), but still, there is a kind of vacuum present in a genre research-article that does not somehow quote Miller. 

So, the article is as influential as such things ever get. However, the article's impact wasn't immediate. It would require a thorough bibliometric tracking to trace its full reception history, but insofar as I have been able to determine, "Genre as Social Action" did not conquer the center stage until after the publication of another crucial work, Aviva Freedman and Peter Medway's anthology Genre and the New Rhetoric. It did have quite a lot of happy readers before that, but Freedman and Medway's anthology seems to mark the turning point after which Miller's study had become so crucial it could no longer be simply "Miller (1984)".

So, the question presents itself: When do you know that a study is high-impact? If you measure a study's impact as something taking place within a certain window, how do you know that you actually have the right window? If high-impact is what you are looking for, then to a certain extent you cannot even know when a study is taken up and suddenly gets hurled towards the academic stardom of the "trailblazing" study?

This is particularly relevant since funding bodies, promotion commitees etc will look to promote researcher's who can make an impact, but often this means that anything older than, say five, years, is counted less because it does not reflect the current stature of the researcher, and its impact is already in the past. 

But don´t trust the administrators. (Quoth the administrator). Research itself seems to be a much more patient endeavor, and in particular within the Humanities the possibility always remains that somewhere a long time ago, somebody wrote just that article that will blaze your trail, break your ground, or be your seminal influence. 

Oh, and if you are interested in narratives, I did a slightly backwards, literary homage to Miller here.