I think it would be possible to reach this conclusion. Using shorter surveys, like Google opinion surveys or something, asking people if they recognize 2-3 logos at a time, run a few hundred of those surveys over a few years and you could categorize each logo based on % of participants who recognized, anything over like 66% could be considered “generally recognizable” and then count how many generally recognizable logos you had.
I think it would be possible to reach this conclusion. Using shorter surveys, like Google opinion surveys or something, asking people if they recognize 2-3 logos at a time, run a few hundred of those surveys over a few years and you could categorize each logo based on % of participants who recognized, anything over like 66% could be considered “generally recognizable” and then count how many generally recognizable logos you had.