Are you still doing phone research? I used to participate monthly back when I was in college.And many of the record industry metrics, whether they be physical or download or stream, do not include information on who was buying or streaming the product. So a "huge" song might be a total negative among a specific station's listeners.
That was the reason why Gavin created his tip sheet back in the later 50's: to allow one format to track the play and response for songs within that format's listeners.
It's why, back in the early 70's I started watching store sales by sending a station rep into a few stores on peak sales days to see who was buying the songs. We did not care how much a song sold; we cared whether our own listeners were buying it.
The 70's spawned many more "tip sheets"* that were specific to certain formats; the use of trades like Billboard and Cash Box were not really used for charts (if stations got them at all) but, instead, for industry and artist news, concert information and other related information. By the end of the 70's we had Radio & Records, later shortened to R&R, which was radio based with industry news for both radio and music.
Today, the source of airplay info is the combination of current "callout" (which is mostly online now) for a short list of songs, and AMT's for library. That tells us what our own format (us and any direct competitor) listeners like, dislike, are tired or burnt out on, are unfamiliar with, etc.
And dissection of test results, such as cluster/factor analysis, shows us how subsets of our own listeners react to songs. We can thus find the songs that have cross specrum commonality, which are the only ones one-to-many media can play.
Actually, thanks for this info as I did not know fully what it research included currently.