Lost in this conversation, I feel, is the use of surveys to learn about the audience (assuming a good sample) rather than naively deciding things for you. I might ask about a topic --- let's say, if D&D should have classes --- not because I wonder that, but because I want to know how that opinion correlates to player experience levels or age groups or whatever. And then maybe I am surprised to learn that class-removers are predominantly DMs, or play in person more than average, and try to explore that.
Surveys aren't always about The Price is Right-style looking back at the audience to figure out what to do. Probably, they should only rarely be that, actually, but I wouldn't discourage surveys overall. I think maybe WotC set themselves up to fail, based on how people see how the process went.