Jon Ericson, then a community manager at Stack Overflow who was trying to get company leaders to repair their relationship with the community, wrote a blog post about the difference between control and agency and how Meta, the "town hall" for the community, factored in. (The company had decided that users on Meta should be ignored because they complained about stuff and, really, they're only something like 0.015% of the community so we can ignore them.)
I left the following comment:
Part of the challenge is that not only do users need some agency, but we need to believe we have some agency. This is a core part of what your employer violated in September; last year they made an explicit promise to moderators (of agency and several other things), and then they grossly violated that in a profoundly hurtful way that they've refused to mitigate at all. Against that backdrop, changes that appear to shut down feedback channels face an uphill battle.
Meta doesn't scale, I know. I'm not active on SO meta where I hear things have broken down. On my smaller sites, meta still works fine -- not only that, but it helps us actively build community! Main meta is somewhere in between -- not as big as SO meta, bigger than any other site's meta. I've seen main meta be very productive and collaborative, and I've seen it be...not. I don't think that's just scale; it's also tied to trust (which is related to and perhaps derived from agency).
It's important, therefore, to not just say "meta works" or "meta doesn't work" but to drill into what characteristics make it work or not work. You indicated that scale is an issue. Here are some others:
Visibility. People feel they have agency when they can see that their concerns have been raised (or that they've been able to raise them). Meta is default-public; blog comments (on the Stack Overflow blog) are default-private; SurveyMonkey results are private. Users don't know what's going into that black box.
Responsiveness. Yeah, I get that some employees are afraid of meta. The CMs never seemed to be and you're supposed to be a bridge to and from the community, so it feels like the company could still engage with users on meta if the higher-ups wanted to. People don't expect to control things but do expect to be heard and be able to get answers. The company has been failing badly here, and a move to either shut down or officially ignore meta would make that worse. Y'all could mitigate it by reporting and responding to that black-box survey feedback you're getting, but it'll probably be too little too late too vague. Your users have poured heart and soul into your sites for (sometimes) 11 years; it's very alienating to be shut out without any acknowledgement of the value we've brought.
Gratitude. Sure there are some cranks on meta, but the vast majority of users who are requesting features, raising discussion points, and asking questions are doing so because we're trying to make our shared space even better. We care. We care enough to spend hours developing and presenting ideas, even building prototypes or mockups, to help other users and thus the network. Are some of those offerings crap? Sure. But the company shouldn't be so quick to dismiss these offerings anyway. The person who wrote that crappy memory-hogging userscript today could build a really thoughtful spec to solve a pain point tomorrow -- if you don't send the "don't bother, kiddo" signal. There has been too little acknowledgement of and thanks for the many contributions the meta community has made over the years. Instead we're branded as malcontents or worse and dismissed as dangerous for employees' wellbeing.
I know that you, Jon, know all this already. But sometimes even when I know something, hearing somebody else say it helps me see it in a new way that then helps me act on it. You know all this, but perhaps, maybe, my commenting on it anyway helps you do something with it to get through to the people who don't know it and seem uninterested in listening to mere users.